Final Report Summary - ABC4EU (ABC GATES FOR EUROPE)
Executive Summary:
ABC4EU project started in January 2014 to address the topic SEC-2012.3.4-6 Enhancing the workflow and functionalities of Automated Border Control (ABC) gates, in the context of the 7th Framework Programme.
Before the kick-off of the project, many ABC gates systems were deployed in the main European airports, not only as pilot projects intended to test the capability of these systems to improve the border crossing processes in aspects such as speed, security, process automation, false rejection reduction, etc.; but also as fully operational deployments, designed to facilitate the border crossing process to bona-fide travellers while maintaining the security at the border.
One of the issues to be addressed by ABC4EU project has been the need for a harmonised approach in areas as e-passports management, biometrics, gate design, human interface, processes, PKD certificate exchange, signalling and interoperability.
On the other hand, the EC published in 2011 a new approach for border management which implied an impact in ABC systems development (Smart Borders initiative, formerly known as Stronger and Smarter Borders), including an integrated, EU-wide EES mainly aimed at extending the use of automated systems to third country nationals, while helping identifying irregular immigration, apprehending overstayers, and generating strategic information to fight and prevent criminal activities. In the scope of the initiative, a Registered Traveller Program (formerly known as National Facilitation Program) was conceived to enhance automation of the border checks, aimed at frequent, pre-screened and pre-vetted third country nationals crossing the borders.
These topics were addressed in the different stages in the project:
• Assessing the impact on the privacy and legislation of the project activities
• Analysing the state of the art at the technical level, as well as lessons learnt of previous pilot and operational projects deployed looking for further automation
• Designing the evolution of the systems to be adapted to the changing legislation
• Setting up the overall architecture of the subsystems developed
• Adapting the systems to achieve the goals of the project
• Executing pilots to assess the technical developments in two stages, one for testing prototypes and the second for final systems, in operational conditions
• Ensuring proper dissemination in workshops, seminars, etc of the activities and outcomes of the project
The impact of the project development at the privacy and personal data management dimensions was early anticipated and addressed since the very beginning of the project via a transversal activity, by defining a framework for privacy management, training and continuous overview of the activities of the project and assessment of the outcomes of the partners’ activities during the project execution.
Project Context and Objectives:
The project objectives were in line with the topics and issues to be addressed in the call SEC-2012.3.4-6 Enhancing the workflow and functionalities of Automated Border Control (ABC) gates:
• The development of a Smart Borders initiative includes establishing a central Entry-Exit System and a Registered Travellers Programme which facilitates border crossings of TCNs.
• The ABC Systems currently deployed at some of the external borders regularly uses only one biometric identifier (face or fingerprints)
• Can only be used by EU citizens holding e-Passports or National ID cards.
• The future EU policy will require the processing of TCNs at the external borders by verifying their biometrics through automated processing (ABC eGates).
• UE citizens will also be able to make use of second generation e-Passports including multiple biometric identifiers.
• ABC systems implemented by some Member States allow for automated border checks based only on facial image verification.
• Moreover, the user interface is not being harmonised among Member States.
• This pilot should address these two issues: to propose a harmonised and coherent solution, allowing using the e-Passports to their fullest potential during the automated processing.
• Attention should be given to compliance with European societal values, citizen’s rights including privacy and acceptability.
ABC4EU project consortium was originally composed by 15 partners from the industry, academia and end-users from 8 EU countries, and changed during the project execution. Two partners left the project during the first year, one industry partner joined to take over certain activities and a fifth end-user engaged with the project at the later stage for supporting diversification of border crossing points and countries where the pilots were deployed and validated.
The main goals of the project were:
• Analyse the ABC concept vs EU legislation, in particular the Schengen Border Code, to identify which processes can / cannot be automated and the impact that different levels of automation could have in the SBC procedures.
• Optimize and harmonize the ABC processes related to the increasing flow of people crossing borders, without undue delay and with minimal intrusion.
• Assess the feasibility of a future National Facilitation Program (NFP) and the Entry / Exit System (EES) in the EU which could be integrated into the ABC gates from technical, functional and ethical point of view.
• Evaluate, optimize and harmonize the use of 2nd generation passports biometrics (facial and fingerprint) for passengers identity verification.
• Identification and design of a basic set of features to achieve a common user experience, involving ABC infrastructure, passenger verification processes, interface look and feel, signalling, etc.
• Identification and, to the extent possible, overcoming of obstacles which are preventing today the certification distribution throughout Europe to access the fingerprint information in 2nd generation ePassports.
• The compliance of the proposed processes with the legal requirements such as respect for fundamental rights, in particular the protection of personal data and with the ethical principles.
• Design of a C4I concept for regulated borders, integrating EES, NFP and ABC gates.
ABC Gates
The activities in the scope of ABC Gates redesign were mainly defining a harmonized approach in areas as e-Passport management, biometrics, gate design, human interface, processes, PKI certificate exchange, signalling and interoperability.
Ensuring the integration of multiple biometrics in the identity verification process, as included in 2nd Gen e-Passports, and to allow third country travellers registration in the EES and NFP.
Anti-spoofing developments, focusing on both biometric identifiers, face and fingerprints, which will allow for enhanced security.
The development of new support systems in line with the Stronger and Smarter borders regulation, fully integrated in the ABC Gate concept.
2nd generation passport
Setting up the technical features to maximize the use of 2nd generation e-Passports to their fullest potential.
Trying to overcome the obstacles that prevent the certificate distribution and efficient travel document inspection systems.
Implementation of multimodal biometric fusion during the identity verification process, while the convenience for the traveller is increased, as different biometric markers are available.
Next generation biometrics
Research on new anti-spoofing techniques, for both fingerprint and facial biometric markers.
Development of multimodal biometric fusion algorithms to efficiently address identity verification of the traveller, aimed at both stationary, regular ABC deployments and mobile devices.
Balancing security/convenience by defining:
• A “passenger experience” oriented scenario with face as main biometry and fingerprint as a fall-back verification when applicable, with faster clearance process cycles.
• A more stricter multimodal fusion scenario for reinforced security, with lengthier clearance process cycles.
Stronger and Smarter borders integration
ABC4EU has developed a National Facilitation Program system, in line with the latest trends of Smart Borders package, aimed at facilitating the border crossing of frequent, pre-screened and pre-vetted third country nationals.
Many of the entry – exit records for the third country nationals have been automatically registered in the ABC Gates, therefore, ABC4EU proposal assumes that the analysis of the practical feasibility of this initiative is been also included in this project.
EES and NFP prototypes have been developed and integrated with several ABC gates in different countries in order to test and validate its feasibility from the legal, functional, technical and ethical perspectives.
Mobile solution prototype
Different kind of mobile devices available in the market, aimed at different user segments, were analyzed to develop an efficient solution for border checks in mobile scenarios.
The aim of the project has been to develop a prototype, handheld-based solution, possible to be used inside trains and buses, while the Border Guard is in motion, covering all border scenarios.
Pilots
The sites for testing have been Portugal (airport), Spain (airport and seaport) and Romania (land border). The aspects of the project that were validated:
• Ergonomics for the operational End User and traveler.
• Document verification (physical and logical) capabilities.
• Biometric identity verification (per type of biometric marker and verification platform/hardware).
• Integration with external systems (certificate exchange, VIS, SIS, EES, NFP).
• Mobile/integrated solutions.
• Central control and support systems.
The pilots were executed in two stages, with the objective of testing a first prototype to be redesigned and reengineered into a final system to be tested in the second pilot round.
Engineering and societal impact
The impact on the ethical and privacy dimensions was early anticipated, even before the project started. In order to address these topics, dedicated ethical and societal review partners established the early inclusion of legal framework at the beginning of the project, including targeted training for designers/engineers to ensure privacy protection and societal awareness by design and ongoing review of every work package by the ethical/societal partners.
The project partners implemented a holistic approach to societal impact, including privacy and data protection but also factors relating to social acceptability and ethics to ensure the success of the project.
Project Results:
First period (2014-2015)
During the first period of the project (from month 1 to month 12) the activities were focused on:
1. Legal, Social and Ethical aspects
After an initial assessment of the procedures for Privacy Management, providing a specific set of guidelines for ABC4EU project, WP1 started with the review of the Legal Frameworks, the review of case studies, providing a comprehensive analysis of ABC/EES/RTP systems deployed throughout Europe with a view on some significant cases outside Europe, the anticipation of externalities and social costs of implementing ABC gates and the analysis on the acceptability and feasibility from legal, ethical and social perspectives. By completing these activities, the contours of the legal, social and ethical assessment of the project began to be laid out. Then, the methodology that developers can use when designing and engineering ABC gates was produced and tested for the first time in the assessment of the work carried out in the scope of ABC Processes and Workflow design.
2. Technology Assessment and End Users Requirements definition
The main objective of the WP2 is to provide the Consortium with a set of common functional requirements for ABC systems from the End User perspective. To achieve this objective, the WP2 defines a series of activities which are aimed to:
• Get the state of the art of technologies which are potentially applicable to Border Control.
• Identify mandatory and forbidden checks to assure the specific functional needs and requirements of End Users at BCPs are performed according to law, covering all types of BCPs.
• Determine the specific functional and information needs and functional requirements at the strategic level for Border Guards, for national and supra-national coordination centres.
• Understand the needs and requirements set out by pan-European initiatives and systems, including existing (SIS/VIS and API) and prospective ones, such as the EES and the RTP.
• Obtain the experience stemming from those Border Guards who already operate ABC systems and biometric-based border controls.
• Understand automation possibilities through an analysis of automation VS security trade-off for the different types of travellers, including those already defined in the Schengen acquis (EU/EEA/CH citizens and Third Country Nationals) and those which would be defined as part of new EU initiatives, such as an RTP.
3. ABC Processes and Workflow design
The global objective of this Work Package is driven by the need of harmonization of ABC systems, in terms of traveller experience, speed up traveller flows and optimize solutions of ABC e-Gates. To achieve this goal, the activities were focused on:
• Define a common architecture for the design and operation of ABCs (e.g. the configuration, operational rules, interface with the travellers, and intervention of the border guard personnel).
• Assess the technology evolution of the biometric data in the passports, including an analysis to the applicability of new features of 1st and 2nd generation of passports.
• Evaluate and assess the usage optimization of the 2nd generation of EU eID cards and residence permits, as well as, the Registered Traveller Token scenarios in order to provide recommendations for Third Country National Unique Identifier in terms of performance, usability and economic viability
• Evaluate and determine the interoperability requirements of a BCP with external subsystems, with a concern in harmonization while designing the corresponding interfaces.
WP1 LEGAL, SOCIAL AND ETHICAL ASPECTS
Objectives for the period
Within WP1, the theoretical and practical elements of the project’s societal impact assessment were carried out:
• It was provided a legal mapping of the EU’s border management and privacy & data protection frameworks, as a means providing project members an understanding of the regulatory framework of ABC.
• Societal impact questions were anticipated by reviewing previous cases of ABC in the EU and beyond, with emphasis also placed on entry-exit systems (EES) and registered traveller programmes (RTP) that might provide best practices for the EU’s planned harmonized implementation of these systems.
• It was provided background on the stakes of border management from the standpoint of questions relating to acceptability, desirability, ethics and data management.
• It was provided a methodology for making other project members aware of societal impact, and provided a practical basis for training relating to legal and ethical questions.
Summary of the tasks carried out and significant results
Activity 1.1 Review of Legal Frameworks regarding technologies and border management
This activity has carried out a comprehensive analysis of the relevant legal framework at both EU and Member States (selected case studies) levels.
Activity 1.2 Review of Case Studies
The aim of this activity was to study the earlier experiences with automated border management so as to lay out the previous challenges faced by those systems, and to anticipate the potential legal, ethical and social externalities of the mismanagement of specific cases.
Activity 1.3 Acceptability and Feasibility from legal, ethical and social perspectives
This activity reviews relevant literature as well as lessons and suggestion from different civil society groups and bodies related to technologies applicable in the scope of ABC4EU in order to start working on a methodology to address ethical and social concerns in the project.
Activity 1.4 Anticipation of externalities and social costs of implementing ABC Gates
This activity explored the interaction of guaranteed rights and device functionality and lay out impact maps at the legal, ethical and social levels, taking a further step in the shaping of the methodology.
Activity 1.5 Methodology
This activity brought together the knowledge produced by the previous tasks in order to produce the methodology that developers can use when designing and engineering ABC gates.
WP2 TECHNOLOGY ASSESSMENT AND END USERS REQUIREMENT DEFINITION
Objectives for the period
The objective is to provide the Consortium with a set of common functional requirements for ABC systems from the End User perspective, not only providing an adequate framework for following WPs, but also serving as a baseline for harmonization and standardization.
Summary of the tasks carried out and significant results
Activity 2.1 Technology Assessment
This activity is aimed to be a survey of the state of the art in the existing and emerging technologies which may have an impact on Automated Border Control systems. Furthermore, the results obtained at the end of this survey are intended not only to lay down the foundations for future decisions about which technologies should be taken into account in the next stages of the ABC4EU project, but also to provide information to determine the actual feasibility of End Users’ requirements. This activity culminated with the main result of the deliverable 2.1 “Technology state of the art survey.”
Activity 2.2 Impact of EU initiatives
This activity is aimed to evaluate the impact of the integration of current EU initiatives in the Border Control Point process and aid to determinate the End-Users requirements according to current European legislation. This activity culminated with the main result of the deliverable 2.2 “Impact of EU Initiatives.”
Activity 2.3 ABC experience. Lessons learnt.
The activity reports on the lessons learnt and best practices in automated border gates obtained from Border Guards and from initiatives from EU, in particular Frontex, and other relevant entities, for example IATA. This activity culminated with the main result of the deliverable 2.3 “ABC experience. Lessons learnt”.
Activity 2.4 Border automation analysis
It is aimed to analyse the activity of crossing a borderland through a Border Control Point (BCP) from a point of view of what sub-process within BCP process are susceptible or are non-susceptible of automation. This activity culminated with the main result of the deliverable 2.4 “BCP process. Automation Analysis”.
Activity 2.5 Study of traveller flows in BCPs (airport, land and ports)
The purpose of this activity is to study traveller flows in different type of Border Crossing Points (BCP) to define guidelines for the harmonization of automated. This activity culminated with the main result of the deliverable 2.5 “Study of traveller flows in BCPs (Airport, land and ports.)”
Activity 2.6 EU common basic set of functional requirements
This activity is aimed to collate and review a common set of requirements with EU-wide applicability, focused on the functional needs of the operational End User at the BCP (be it air, land or sea). This activity culminated with the main result of the deliverable 2.6 “EU common basic set of functional requirements.”
WP3 ABC PROCESSES & WORKFLOW DESIGN
Objectives for the period
The global objective of the WP3 is driven by the need of harmonization of ABC systems, in terms of traveller experience, speed up traveller flows and optimize solutions of ABC eGates. The activities provide to the Consortium a set of processes, workflows and recommendations for the ABC design from the theoretical and practical Consortium partners’ experience, serving as a consolidation of the available ABC knowledge and a baseline for the subsequent work (analysis, design and development WPs).
Summary of the tasks carried out and significant results
Activity 3.1 Harmonized rules for ABC design: Processes, Human interface, signalling
This activity is aimed to define a common architecture for the design and operation of ABCs: the configuration, operational rules, and interface with the travellers and intervention of the border guard personnel. This activity encompasses the following tasks:
• Gathering and description of a common set of business rules which could be applied through all the Border Guard Authorities in order to define a harmonized logic.
• Description of the available signage at airports and during ABC usage (e.g. media options, human factor, usability) in order to reach at a set of recommendations and best practices.
• Assessment of the already implemented ABC designs (e.g. single-step vs integrated two-step vs segregated two-step process, single door vs mantrap configuration, swing doors vs sliding doors).
• Assessment of the specific characteristics of travellers with special needs (e.g. travellers with reduced mobility, visually impaired travellers, minors) and their relationship with the ABC
• Assessment of the streamlining of the human interaction (e.g. travellers with the ABC, border officers with the ABC, travellers with border officers).
• Evaluation of the ABC design with the objective of optimization for facial recognition.
This activity culminated with the main result of the deliverable 3.1 “Guidelines for technological harmonization of ABC: view of the end-user”.
Activity 3.2 Biometrics: Evaluation and optimization of 2nd Gen ePassports
The aim of this activity is to assess the technology evolution of the biometric data in the passports, including an analysis to the applicability of new features of 1st and 2nd generation of passports. This activity encompasses the following tasks:
• Assessment of the quality and efficiency of the passports, specifically the face image quality for facial recognition, fingerprint image quality for fingerprint recognition, passport performance (e.g. security features, average times, electronic authentication) and passport improvements.
• Description of a multi-modal biometrics model and assessment of its advantages, performance, usability and convenience for biometric verification.
• Assessment of biometric passport quality from different countries, through the development of a Gap analysis based on procedures and biometric data specifications (survey analysis) in order to define a set of recommendations to optimize facial and fingerprint recognition.
• Analysis of both ABC and passport working together in order to increase convenience and performance for the traveler, focused on electronic document reader, face image and fingerprint image improvements.
• Assessment of the traveller data protection theme.
This activity culminated with the main result of the deliverable 3.2 “Assessment of the compliance of new biometric data supports for ABC - 2nd Gen ePassports”.
Activity 3.3 Biometrics: Evaluation and optimization of 2nd Gen EU eID cards and residence permits
This activity is aimed to assess the usage optimization of the 2nd generation of EU eID cards and Residence Permits, as well as, the Registered Traveler Token scenarios. This activity encompasses the following tasks:
• Study and analyse the compatibility of eID cards and Residence Permits with ABC, through a Gap analysis and assessment of the different document specifications, in order to provide conclusions and a set of recommendations.
• Evaluation of the face and fingerprint image quality stored in both documents.
• Assessment of the different Registered Traveler Token scenarios, through the usage of three different elements to compare them: performance, usability (from a traveler’s and border guard’s perspective) and economic viability.
• Description of procedures in order to define a set of recommendations to optimize facial and fingerprint recognition.
• Assessment of the societal impact context for the deployment of eID cards, residence permits and the proposed use of Registered Traveller (RT) tokens in the Smart Borders package.
This activity culminated with the main result of the deliverable 3.3 “Assessment of the compliance of new biometric data supports for ABC - 2nd Gen EU eID cards”.
Activity 3.4 Interoperability of ABC points with external systems
The activity 3.4 is aimed to evaluate and determine the interoperability of ABC points with external subsystems, with a concern in harmonization while designing the corresponding interfaces. This activity encompasses the following tasks:
• Description of the different systems that compose an integrated solution for BCPs within the Schengen area and assessment of the interoperability requirements inherent to a BCP, considering both existing and new subsystems.
• Assessment of the horizontal integration of ABCs with border management systems.
• Discuss the technological aspects that should be taken in consideration when implementing the solution: interface technology, software architecture, communication flow and data security.
This activity culminated with the main result of the deliverable 3.4 “Interoperability assessment including C4I”.
WP6 BCP Test & Validation
Objectives for the period
The design and development of a validation framework has been the main goal.
Summary of the tasks carried out and significant results
A research was carried out on this stage with regard to the technical, legal and operational levels, with the final aim to find a set of common criteria, and in this way develop a validation framework of all activities involved in the previous WPs. The structure is in line with relevant governance frameworks and standards in order to provide guidance to top management in setting an automated border control system from a governance perspective.
This deliverable is composed of different parts that involve all data required. The preliminary design and specifications of every step in the validation framework have been described, focusing in a high level structure to define the framework. The report will show the results of the task carried out previously, and define the addressee and the regularity of the reports.
Second period (2015-2016)
WP1 LEGAL, SOCIAL AND ETHICAL ASPECTS
The main objectives regarding the legal, social and ethical aspects for the period were the following:
▪ Completion and submission of D1.7 “Assessment of project development from social, legal and ethical viewpoint: WP3 Report”
▪ Completion and submission of D1.8 “Assessment of project development from social, legal and ethical viewpoint: WP4 Report”
Summary of the tasks carried out and significant results
Activity 1.6 Assessment of social, legal and ethical compliance during project development
Within WP1, the mappings developed in early deliverables, D1.1 to D1.4 and the methodology first laid out in D1.6 have paved the way and provided a framework for the assessment of WP3, whose results have been published in deliverable D1.7 at the end of February. This assessment has identified a number of key issues that will require further analysis in the forthcoming project phases (namely WP4 to WP6), such as, among others, the scarce evidence provided so far of security, financial, and convenience gains, and consequently, the need for better tools to assess and produce credible data, the lack of a specific legal basis for automated border checks, the problem of non-systematic checks for EU citizens, and some privacy and data protection issues.
All previous deliverables corresponding to this WP (D1.1 to 1.8) have been also considered in the design and development of the pilot test. The WP team has carried out the following tasks:
• Some findings of the ABC4EU project so far, especially regarding the design of the pilots, were presented by Eticas at the EAB Research Projects Conference (EAB-RPC) held last September, 2016 in Darmstadt and organized by European Association for Biometrics (EAB) in cooperation with Joint Research Center (JRC) and Fraunhofer IGD. The name of the presentation was: “ABC4EU pilots: evaluating ABC gates from a legal, ethical and societal standpoint”.
• ABC4EU pilots: evaluating ABC gates from a legal, ethical and societal standpoint.
• Gathering and classification of information about the ethical, technical and societal issues concerning the intervention of the different partners in the pre-pilots. A questionnaire elaborated by Eticas with this purpose and a table concerning the potential use of personal data within the pre pilots was shared with the consortia, and useful inputs were subsequently received from each partner involved.
• Gathering and analysis of information regarding the ethical implications of the ABC gates, following the specifications provided by the consortia partners. Knowledge and results from previous EU projects, such as FastPass, FRA, have also been taken into account.
• Ethical assessment of the data collection tools to be used for the pre-pilots.
• Development of the consent form for the Spanish pre-pilots, with the help and support of Indra.
• Development of a methodological design for the pre-pilots, including the structuration of the variables and the 32 indicators. These elements were incorporated into the survey and a template for non-participant observation was elaborated. Finally, the pre-pilots survey was elaborated and subsequently adapted to include questions and concerns proposed by partners.
• Testing of pre-pilots methodological tools at each location, both Lisbon and Madrid, applying a total of 25 questionnaires and conducting observations at both airports.
• Recommendations regarding ethical and legal compliance of different partners’ participation during the development of the project
WP2 TECHNOLOGY ASSESSMENT AND END USERS REQUIREMENT DEFINITION
Objectives for the period
The objectives and methodology for D.2.7 were successfully established
PwC worked in the review of the contents and the structure of deliverable 2.7 to facilitate the comprehension of the ideas expressed and favour the acceptance of the ABC system, obtaining feedback from the different end users.
After the revision of the structure of D2.7 PwC worked in the definition of a structure for supra-national coordination and made an assessment about the needs for strategic analysis and mechanisms to share information among BCPs and the National Back Office, and between the National Offices and the EU Coordination Centre.
Moreover, PwC established the methodology used to analyse the stakeholders influence and interest on the project, taking into account the workshops carried out with various partners, End-Users and government agencies.
On the other hand, as D2.7 should achieve the objective of standardization, PwC established a common set of guidelines for the coordination and harmonisation among the Governance Board. At this point in this deliverable the SNA methodology, defined by MIR-DGP, and developed as a third dimensional model by URJC, was explained as an important tool to facilitate the understanding of the network´s behaviour.
Summary of the tasks carried out and significant results
Activity 2.6 EU common basic set of functional requirements
The main result of WP2 has been the establishment of the D.2.7 requirements structure and the lessons learnt through our research and the Technical Boarding Meeting with the Consortium members. The consortium has been provided with the following results:
▪ Strategy and coordination requirements based on Governance, Security, Communication, Data Processing, and Legal, Social and Ethical concerns.
▪ Stakeholder analysis
▪ Literature review to validate the requirements
▪ Comparative results using interview skills
▪ Stakeholder analysis.
▪ Literature review to validate the requirements.
▪ Comparative results using interview skills.
▪ Strategy and coordination requirements based on Governance, Security, Communication, Data Processing, and Legal, Social and Ethical concerns.
▪ SNA methodology for a better understanding of networks.
The creation of a new board, called EEUAB (External End-User Advisory Board) was setup to facilitate the cooperation among end users that are not working for the ABC4EU project.
After the revision of the D2.7 content by EC, PwC worked in the inclusion of all corresponding changes that applied. The set of changes applied to D2.7 are described below:
▪ Clarifying the sources of the requirements.
▪ Clarifying the participants of the surveys used.
▪ Adding new literature review.
▪ Including critical assessment taking into account the referred frameworks.
▪ Modifying the deliverable with a specification of the methodology (who decided the requirements and how).
▪ Including the specific objective and tangible impact of that work. Specifying which its contribution to the project was.
Under Activity A2.6 the project partners established end-user driven functional requirements for the National Facilitation Programme and Mobile subsystems workflow through collaboration with Spanish CNP Border Guard police. The need for seamless integration in a harmonized ABC4EU border-crossing approach was a central requirement needed for designing the systems interoperability to be deployed by multiple Member States in heterogeneous BCP settings. The key goal of facilitating TCN passenger border crossing flows, was updated to follow the most recent EC proposal for the revised 2016 Smart Borders package and clarified the maximum levels of privacy and security so to align with the EU data protection regulatory framework including the GDPR that concludes all biometric data that can re-identify a person is classified as sensitive data, while enforcing the social and ethical aspects from WP1.
URJC team worked in deliverables 2.6 and 2.7 in the development of a methodology regarding the Social Network Analysis (SNA). This methodology brings forward a study to establish a model and describes the structure of the End Users community from the SNA point of view by presenting the results in a matrix of links between nodes. Different cases of simple networks for the end users have been studied and performed in the methodology as the annex shows, but in this deliverable we have chosen the cases that shall apply. Moreover URJC has considered three authority levels with many interconnections in the global network. Different virtual graphical representation of the cases studied were included.
WP3 ABC Processes & Workflow design
Summary of the tasks carried out and significant results
Activity 3.3 Biometrics: Evaluation and optimization of 2nd Gen EU eID cards and residence permits.
During the analysis phase, the following goals were achieved:
▪ Completion of assessment regarding Ergonomics and Biometrics Needs in the case of facial biometrics. All the different physical configuration was analysed, and several recommendations were made in each configuration.
▪ Completion of assessment regarding ABC for passengers with disabilities in the case of facial biometrics.
▪ Completion of assessment regarding Streamlining human interaction..
▪ Completion of assessment regarding the challenges that must be addressed when improving face image quality for biometric recognition. An experimental evaluation with real data, evaluating the influence of the illumination over the final facial image quality in an ABC environment was presented.
▪ Completion of assessment regarding the application of multimoda fusion in the case of the ABC systems.
▪ Completion of assessment regarding the state of the art in the case of the research about face image quality for biometric recognition.
WP4 EU ABC ARCHITECTURE
Summary of the tasks carried out and significant results
During the analysis phase, the following goals were achieved:
▪ A very high level (from the point of view of the traveller) of the different processes that a traveller must undergo when interacting with each subsystem and high level data entities involved in those processes for better understanding were streamlined.
▪ Requirements Catalogue: With the information generated in previous deliverables and the decisions made in the Business Process Model, the main requirements that an ABC System should address were identified, not only inside the scope of the current project, but also for their consideration in future European R&D projects
▪ Detailed model were analyzed, studying the system under different perspectives:
• Business perspective (use cases model)
• Data structures perspective (data model)
• Functional perspective (class model)
• Interactivity with travellers perspective (user interface model)
For the preparation of the pilots, the work to be done was defined in several business cases, including
▪ Graphical location of the pilot into the general process of each subsystem
▪ Definition of the scope of each pilot by narrowing the requirements identified in previous analysis, while extending the requirements catalogue with specific needs for each pilot
▪ Description of work: clarification of what outcomes are expected at the conclusion of the project, and how they would be achieved.
▪ Design models for those pilots that will implement software modules, including new point of views of the system:
• Dynamic point of view (scenarios/ State-Transition diagrams)
• Data model
• Class model – component and deployment Models
• User interface model
The design of the architecture for the different subsystems (Biometry, E-documents, RTP, EES CES, ABC interoperability, external systems interface) was performed according to the requirements. A common BMS (Biometric Matching System) for the EES and the RTP was designed, and the interaction among the different subsystems during an ABC crossing by a TCN was analysed and defined in the architecture definition. Also a C4I Integration Concept was written, which analyses how such a system could be integrated with an ABC system.
The deliverables were adapted to the requirements of the new Smart Borders proposal (released April 2016), as it affects the ESS and the RTP (which becomes NFP) mainly as the project commited with adapting the work to the everchanging legislation.
Additional research in the scope of WP4 carried out by the consortium:
▪ Analysis of current and state-of-the-art biometric solutions implemented or potentially applicable to the ABC Systems, VIS, EES and RTP. Focus on facial and multimodal biometrics.
▪ The list of functional requirements in an ABC system related to biometrics module were proposed.
▪ Analysis of the current potential weakness of the biometric module in ABC systems. Two main fields to improve the current state of the art were detected: multimodal fusion and presentation attack detection.
▪ Development of two business cases according the weakness detected: multimodal fusion and presentation attack detection.
▪ Proposal and guidelines generation of the biometric module High Class Model (logical model of the software system under construction).
▪ Several guidelines for the Signalling User Interface Model were developed considering different international standards.
Because of the general last years increment of the cyberattacks in all IT systems, the effort dedicated to antispoofing or presentation attack detection techniques has been bigger than the initially estimated in the project proposal. The cybersecurity field has suffered a great expansion during this last years at the same time the attacks are more and more complex. There are multiple examples of attacks performed during last years that was absolutely unexpected several years ago. A concrete proof of the increment of this technical necessities is the currently in process development of a new ISO standard (ISO/IEC CD 30107) focalized specifically in presentation attacks.
From the research point of view this field has not been deeply analysed in the case of ABC systems, so the project results were adapted to current social necessities.
From a technical point of view, a wider range of attacks than the initially expected and with higher technical levels were considered: printed face on a paper sheet (as baseline method); digital face displayed on a screen from digital devices such as tablets, smartphones, and laptops; 3D masks (paper, silicon, cast, rubber etc.) specifically moulded for a targeted face, inclusion of facial data in alternative poses (like t-shirts), etc..
The inclusion of these more complex attacks has consequently resulted in the necessity of the analysis, design, development and test of more complex presentation attack detection methods. URJC research team has included at the same time nor only more sophisticated algorithms, but already a multisensor (and multispectral) approach in order to study the antispoofing capability of this different visual information representation.
WP5 SUBSYSTEMS DEVELOPMENT & INTEGRATION
Objectives for the period
The scope of the pilots was drafted in order to anticipate the research and development activities within WP5.
The objectives in WP5 for this period were the following:
▪ Development and completion of deliverable D5.1 Biometric proof of concept.
▪ Development and completion of deliverable D5.3 Travel document management proof of concept.
▪ Development and completion of deliverable D5.5 RTP subsystem proof of concept.
▪ Development and completion of deliverable D5.7 EES subsystem proof of concept.
▪ Development and completion of deliverable D5.9 ABC Systems proof of concept.
▪ Development and completion of deliverable D5.11 Mobile subsystem proof of concept.
Summary of the tasks carried out and significant results
Within WP5, the Technical and Logistical aspects of the Proof of Concept in Lisbon and Spain were carried out.
Activity 5.1. Biometrics
The architecture of the biometric software provided by the partners was analyzed, in order to be integrated into the enrolment and verification processes.
The upgraded ABC System deployed in Spain started to be designed and built, with new camera system to support the research activities. A preliminary software module for retrieving and processing the video stream was developed by INDRA.
A comprehensive study of different algorithms for object and persons detection was performed by INDRA using the 3D sensor located over the eGate (mantrap) in a zenithal position. Tailgating and abandoned objects detection algorithms were developed. For detecting persons, head and shoulders are searched taking into account the difference in depth of head and shoulders and their apparent sizes. In order to detect abandoned objects, the 3D image is processed after noise elimination and background subtraction. The algorithms were successfully tested during Barajas pilot in numerous situations.
Atos proposed a common SOA architecture for RTP and EES to make use of a common BMS provided by the Dermalog AFIS. The AFIS would store fingerprints and perform biometric matching for fingerprint queries and it would store the facial images to be retrieved by ABC systems which would perform the matching locally.
For the biometric prototype Cognitec started research activities in order to improve biometric performance of face recognition.
In particular, investigations concentrated on the border control use case (camera versus passport image) with passport images of reduced quality, as they were reported. Apart from general improvements of the biometric performance in terms of False Acceptance Rate and False Rejection Rate, Cognitec was able to achieve improvements especially for images taken with imperfect lighting conditions and non-standard pose.
The results of these developments were provided to the partners in the project by means of the Cognitec product FaceVACS SDK v. 9.2.
To further improve the prototype biometric algorithms Cognitec addressed problems like compression artefacts in passport images and partial facial occlusions including glasses, which can occur with both camera and passport images. Cognitec was able to achieve some improvements here, though these are not yet consistent on all types of images present in the test bench. So, this development was considered still ongoing and required a later update.
URJC team has worked in the study and development of different Presentation Attack Detection (PAD) techniques. Different methods to detect spoofing attacks in the case of facial biometrics were analysed. After a deep analysis of the possibilities available, four types of vision sensor (RGB, IR, NIR and depth information) were selected and its application to presentation attack detection were studied.
In order to perform this research, the different antispoofing facial databases available in the research community were analysed. One of the main conclusions after this study was that currently any of these databases could apply to an ABC system particularities. Any of them consider the use of a multispectral approach, and the attacks weren’t usually adapted to an ABC environment. Considering this restrictions, a new database called Frav-Attack database was acquired, adapting the environment and the quality of the attacks to an ABC environment.
The acquisition of this new database from our Facial Recognition and Artificial Vision (FRAV) group in URJC wasn’t initially included in the proposal, but after the analysis of current state of the art attack databases, a lack of attack situations admissible for an ABC environment was detected. The final decision was to design and acquire this new FRAV-Attack database to cover this gap of quality data. It is an extra-effort but this database is highly relevant for research and experimentation in this project, and globally it is a significant contribution.
On the other hand, URJC team researched in the development of optimal Multibiometric fusion techniques for ABC gates. The methodology to analyse specific parameters in ABC systems (nationality, age, date of issue of documentation, etc) and how they can affect biometric subsystems were developed.
For the biometric prototype Dermalog focused on increasing the customisation possibilities and stabilisation of interfaces. Additional effort was brought to the automation of the Web AFIS installation.
The results of these developments were provided to the partners in the project.
Dermalog’s “Web AFIS” biometric prototype represents an AFIS (Automatic Fingerprint Identification System) that provides a centralized Web service interface.
The pilots developed by INDRA, ATOS and VISION-BOX for ABC gates and mobile verification solutions connect to the database going through this “Web AFIS“, which then enables access to the required operations on biometric and/or demographic data.
During this period, Dermalog also tackled research activities in order to improve the reliability of fake fingerprint recognition:
▪ significant increase in sample size of real humans’ and fake fingers (including increase in variability of fake materials)
▪ enhancement of fake detection support for optical as well as capacitive fingerprint capturing technology
▪ extension of fake detection to support different fingerprint scanners (up to now ZF1, LF1, ZF10, LF10)
▪ substantial improvements of fake finger detection classification algorithms (significant reduction of FRR and FAR)
▪ integration and presentation of fake detection functionality by means of demo applications
▪ development of quality testing and evaluation procedures for fake detection functionality
The results have been provided for the ABC4EU project as a plug-in for the LF10 scanner.
Parallel research activities focused on improving verification processes in the context of aging, to increase the accuracy of fingerprint recognition for persons under 12 years of age. Studies were carried on fingerprint growth and the persistence of fingerprint recognition when aging.
During the course of the project, Dermalog identified an issue in which users at border controls are often confused with the use of fingerprint scanners and passport scanners, and mistakenly place their fingerprints on the passport scanner and their passport on the fingerprint scanner.
It was decided to tackle it by developing a new multipurpose scanner designed for national registration and automated border control especially suited for integration into ABC Gates, able to capture both fingerprints and passport images on the same scanning surface, with the aim of facilitating end-user experience and increasing efficiency and speed of document and identity checking.
Concrete actions during this period have been:
▪ Development of a prototype for a multi-scanner
• Alpha-Release C und .Net SDK for internal Dermalog integration of the VF1 multi-scanner
• Beta-Release C und .Net SDK for external integration of the VF1 multi-scanner
▪ FBI certification of the VF1 multi-scanner
The multi-scanner offers the following features and characteristics:
▪ Captures fingerprints with highest quality and liveness detection/fake finger detection
▪ Scans and reads passports with full color scan, IR, UV and RFID
▪ Analyzes optical security features of the passport
▪ Capturing of signatures (with additional accessory)
▪ Scans passports, ID cards, photographs, flight tickets, arrival/departure cards and other small documents
▪ Reads RFID-Smartcards
▪ Automated detection of fake passports (optional)
Activity 5.2 Travel Documents Management
A vendor-independent security patterns database and the corresponding engine for validating travel documents was designed, and a preliminary version of the engine was implemented.
Later the CES was implemented according to BSI specifications (TR_03129-2).
We would like to highlight that during the study of the BSI specification, INDRA found what seemed to be an inconsistency in the specification. ABC inspection systems should be able to obtain a DS certificate from an external source if a particular eMRTD EF.SOD does not contain the DS certificate, using the SignerIdentifier found on the eMRTD. BSI specification refers to this possibility and it even defines alternate behaviour in one of its WS operations, GetDocumentSignerInformation(), in order to cover this case, but it incomprehensibly misses the necessary output parameter for returning the DS certificate (BSI specification “typo”?).
In fact, this functionality is recommended by FRONTEX best practices and it was proposed to add it to the ABC4EU CES, because, contrary to what was initially supposed, it is not uncommon to find modern passports without a DS in the chip.
Indra informed the BSI about the inconsistency in their specification and they acknowledged the issue. A conversation with BSI followed in order to discuss possible workarounds.
The CES could be considered to be improved in this way for the Final Pilot to create an online DB of DSCs as it is clear than many passports do not include DSC in the SoD. From piloting mobile metrics Atos observed 53% of ePassports did not contain the DS certificate, so providing access to DS Certificates online would make a big impact on validating eMRTD data.
The eMRTD validation subsystem (TDMS) was completed by INDRA, and used for Barajas PoC in: 1) Indra verification kiosk, 2) Indra EES enrolment kiosk and 3) Atos NFP/EES Enrolment Station.
Regarding the Mobile System, Atos worked closely with the French mobile device supplier to improve the interoperability, stability and functionality of the supplier’s e-MRTD reading and authentication SDK which is used by Atos ABC4EU Mobile prototype app, (including a feature roadmap for implementing Passive, EAC and Active authentication for piloting). Much effort was devoted to perform interoperability testing and troubleshooting by Atos for ePassports from around the world for Passive and Chip authentication. Additionally, the mobile has been implemented to make use of either a local CSCA DB or make use of CES for online validation of the CSCA in Passive Authentication.
Activity 5.3 Registered Traveller Programme subsystem development
Atos started the software development on the prototype RTP Enrolment Station (ES) and back-end RTP service. However, in April 2016 the EC released a revised Smarts Border Package which dropped the RTP central service and replaced it with a voluntary National Facilitation Programme (NFP) for facilitating registered travellers. Atos spent a month analysing and agreeing with partners the impact of the revised package, and all partners agreed in modifying the scope of the RTP delivery to follow the new NFP specification for registered travellers. It was also agreed that Atos would additionally expand the scope of the Enrolment Station so to provide a common EES & NFP Enrolment Station instead of just NFP. An updated EES registration interface was re-implemented.
The NFP development results consisted of a combined EES and NFP Enrolment Station (ES) web application, NFP backend service, NIST Pack backend service, Front End Gateway (FE GW), Kibana Logstash audit & event log system. The ES registers VE & VH TCNs and their biographic and biometric data in EES before proceeding to optionally register the TCN (biographic data) in the NFP registration.
The Enrolment Station integration was carried out with EES and NFP systems for enrolment in these services and also with CNP gateway system as it was needed to authorise the BG access (authentication in pilots under real operational scenarios) and also do background checks before enrolling in NFP. Later technical DB issues in Argos resulted that during PoC piloting the background checks were only carried out for ABC verification and not ES enrolment. The ES integrated the Indra eMRTD reader application for reading and verifying physical security elements of ePassports and Visas.
Activity 5.4. Entry/Exit subsystem development.
The EES was reviewed and redesigned (according to the new Smart Borders proposal).
Activity 5.5. Existing ABC upgrades (Airport, land, sea).
The new legislation proposal was also taken into consideration for the roadmap of the ABC setup to be tested in both the PoC and Final Pilot. The roadmap included the main features to be included in both the Enrolment Station and ABC eGates.
The scope for the different pilots was defined and the usage workflows were specified for the different scenarios and ABC topologies to be covered by pilots (mantrap or two-step segregated, TCNVH or TCNVE, first time crossing or not, registered in the NFP or not, etc.). For each scenario it was analysed how the NFP and the EES would have to be integrated. The software modules in existing Indra ABC system that would require modifications to communicate with NFP and EES were identified. It was also analysed how the current legislation and the current implementation of existing systems affects and constraints the implementation of the pilots. The specific hardware devices for pilot implementation were chosen (tablets, document readers, fingerprint readers, cameras, PCs, etc.).
Atos helped define the validation process of TCNs according to the revised Smart Borders package, with particular focus on the design of TCN workflow for visa exempt and visa holder TCNs so that they could be enrolled (using ES and kiosk) and proceed to be validated at the border by using upgraded ABC solution (or mobile information units). Atos also developed the Front End GW to enable easier integration with different vendor ABC WS and Restful interfaces, and thus supports interface harmonization towards back end services such as NFP enrolment status checks.
A complete ABC software suite was developed (verification kiosk, enrolment kiosk, mantrap and ABC monitor) integrated with VISIONBOX EES and ATOS NFP system. The biometric statistical information to be logged by the system was agreed with URJC and UMIL. The effort was first aimed at the mantrap system to be used in Barajas. Later the extensions needed in the ABC software in order to accommodate Algeciras pilot features (segregated two steps with facial recognition as the token for automatic door opening) started.
The inclusion of a presentation attack detection evaluation was performed during the pilots test. Because of the last years increment of the problems with cybersecurity attacks, the number of attacks was bigger than the initial expected in the proposal.
Activity 5.6 Mobile subsystem development
The mobile development activities consisted of developing an Entry Exit Service verification workflow for reading ePassport data, reading Visa data (including OCR capture with camera, as well as swipe reader, especially suitable for visas not properly affixed to passport), taking fingerprints for matching records in EES (VE) and VIS (VH) checking existing EES records, checking backend national and central DBs including VIS and SIS (via CNP Argos system) before electronically stamping the border crossing in EES or referring to 2nd line (manual booth). Authentication access control is implemented though integration with CNP Argos system. The mobile subsystem additionally checks the TCN´s NFP status so to allow BGs to achieve significant time-savings for NFP-enrolled TCNs (as the NFP pre-vetting checks avoid the need for border guards to carry out SBC thorough checks).
The Mobile subsystem proof of concept implementation was made available. Updated EES functions and interfaces were re-implemented.
Activity 5.7. RTP, EES and upgraded ABCs integration
The following tasks were carried out:
▪ Integration with the initial EES.
▪ Integration of Dermalog decadactilar scanner in Indra EES enrolment kiosk.
▪ Integration of Dermalog unidactilar scanner in Indra verification kiosk.
▪ Integration with Cognitec’s 1:1 facial recognition software.
▪ Integration with Atos NFP RESTful service.
▪ Integration of the CES (BSI interface) with Atos FrontEnd server.
▪ Integration of the TDMS in Atos NFP/EES Enrolment Station.
▪ Integration from the Atos Enrolment station and Mobile System over the Atos FE GW towards the initial EES.
▪ Integration from the Atos Enrolment station and Mobile System via EES with Dermalog BMS via the EES with correct image format.
▪ Integration of Dermalog decadactilar scanner in Atos NFP/EES Enrolment Station.
▪ Integration with Cognitec’s facial capture software in Atos NFP/EES Enrolment Station.
▪ Integration of the Indra iBCS eMRTD desk top application in Atos NFP/EES Enrolment Station.
▪ Integration from the Mobile System over Atos FE GW to the CES (BSI interface).
▪ Integration from the Atos Enrolment station and Mobile System over the Atos FE GW towards with CNP Argos Authentication web service.
▪ Integration from the Atos Enrolment station and Mobile System over the Atos FE GW towards with CNP Argos pasoFrontera web service, and checks to national DBs, VIS and SIS with correct NIST Pack formatting for VIS checks.
▪ Integration from the Atos Enrolment station and Mobile System over the Atos FE GW with the Atos NFP service.
▪ Integration of Mobile System with Coppernic API, IntgeratedBiometrics Fingerprint API and Zebra OCR API.
▪ Integration from the Atos Enrolment station and Mobile System towards kibana Logstash event logging system
WP6 BCP Test & Validation
Objectives for the period
With the advent of the PoC execution, the main objective of the period was the definition of the validation framework.
Summary of the tasks carried out and significant results
The consortium has been provided with the following results:
▪ Validation framework divided into: design, performance, and analysis phases
▪ Limitations to take into account from a technical, ethical and operational point of view.
▪ Literature review to validate the framework.
▪ Comparative results using interview skills.
The validation framework was written in its final draft version where:
▪ A Validation Framework was proposed in order to serve as guidance for the team responsible for carrying out the pilot activities during the ABC system development.
▪ The goals and guidelines with regard to the design, performance and analysis of the phase of the ABC system, regarding processes and protocols, were defined.
▪ The responsibilities for defining, executing and analyzing the results of all validation activities to be performed throughout the project were established.
On the other hand, the consortium worked in the first draft of deliverables 6.2 6.3 and 6.4 which are focused on ABC, EES and RTP concept and validation plans.
The validation activities for each system are related to:
▪ Ergonomics.
▪ Document verification capabilities.
▪ Biometric Identification.
▪ Access control and perimeter surveillance.
▪ Integration with external systems (VIS, API, SIS, EES, RTP, etc.).
▪ Mobile/integrated solutions.
▪ Central control and support systems.
Opinions, experiences and feedback provided in the last EUIWG and EEUAB meetings (9-10th of May) were incorporated in terms of: ergonomic, functional and operational requirements, use cases, EES new regulation from the European Commission, etc.
PwC prepared the structure of validation results deliverables aimed to gather the results, information and conclusions obtained during pilot phase.
Likewise, concerning the PoC, PwC elaborated a set of forms to facilitate the collection of data during the pilot phase. Furthermore, PwC gave support to ETICAS with activities related with travellers’ feedback. These forms were focused on gathering the following main information:
▪ ABC: ePassport placement time, ePassport reading time, time for facial image capture, time for fingerprint capture, time for verification process and attempts.
▪ EES: ePassport placement time, ePassport reading time, time for facial image capture, time for fingerprint capture, time for BG authentication process and attempts.
▪ NFP: ePassport placement time, ePassport reading time, time for facial image capture, time for fingerprint capture, time for verification process and attempts.
▪ Mobile device: ePassport placement time, ePassport reading time, time for facial image capture, time for fingerprint capture, time for verification process and attempts.
▪ Others: rejects, failures during processes, feedback, age range...
Moreover, two EEUAB meetings has been prepared by PwC for the attending (online connection) of several end-users for sharing the approach, results, tests and experience obtained during the PoC.
Atos provided contributions to Activity 6.3 with revision of earlier validation plans to take into account the changes needed for NFP as it superseded RTP, adding mobile system details needed for the mobile validation proposed metrics to be obtained during piloting for NFP Service, Mobile System and Enrolment Station. As one of the main integrator/supplier of piloted systems Atos have participated in and supported on site on a daily basis the validation testing with real end-users, travel documents and TCNs of NFP Service, EES & NFP Enrolment Station and the Mobile System at International Madrid Airport piloting site.
Mainly, the Proof of concept execution activities consisted in inviting TCNs in possession of electronic travel document (VE and VH) with support of pilot partners and AENA staff to sign consent form and enter pilot flows. First step is to be enrolled on EES and, optionally, NFP at the Enrolment station, located near the border crossing booths but in a specifically signalled area sufficiently remote to avoid interference with the non-pilot flow and/or queue, and subsequently border guards invite the TCNs to test ABC4EU border crossing using eGate or Mobile System. During mentioned EES and NFP enrolment, the Atos system was recording the processing times of all the relevant events taking place in the Enrolment Station during the process. The selection of these events was informed previously to PwC (in charge of the validation process) by Atos, prioritizing the complete process of enrolment in EES as the most important metric to be analysed since it is in this step that EES thorough checks are conducted, which would save time for the TCNs in subsequent border crossings.
Mentioned enrolment was carried out by a border guard (BG) from CNP, who follows the screen instructions to validate the travel documents, capture fingerprints (optionally capturing face with reflex camera when passport image lacks sufficient quality) and perform thorough checks. The Enrolment Station of Atos is designed to process both short stay visa holder and visa exempt TCNs, meaning a majority of TCNs could be enrolled and, after successfully finishing this enrolment phase, each TCN was accompanied by the BG to either eGate or Mobile system. There, the border crossing happens and EES is updated from the eGate or mobile, having first validated the travel documents and checked status of EES and NFP records for this TCN.
The first proof of concept validation stage aimed at performing an end-to-end testing of all the involved research activities, fully integrated in a real deployment scenario. The main objective of the validation testing was to assess the results of the third country nationals processing through an automated border control, as it could be implemented in the future, including additional functionalities to the currently deployed systems which were modified for the purpose of executing the activities planned in the project.
The results of the research activities were fully integrated and tested (CES with BSI interface implementation, new TDMS travel document physical security features validation, tailgating detection, lost object detection, new biometric algorithms, EES and NFP). By testing with real travellers in a real border control scenario, where the Smart Borders regulation was not approved, the systems were adapted to accommodate to the production police systems and to comply with the currently legislation in force.
The expected outcome of this preliminary proof of concept was:
• Assessment of the interoperability of all the elements deployed,
• Assessment of the impact on the current processes of third country nationals enrolment at the borders,
• Assessment of the impact on the current processes of third country nationals automated processing,
• Assessment of the suitability of a NFP functionality to fully automate thorough checks of third country nationals,
• Assessment of a mobile device aimed at helping the border guard for manual processing,
• Assessment of a new travel document physical security features validation engine,
• Assessment of new algorithms for tailgating detection, both in one step and segregated two steps topologies,
• Assessment of new algorithms for lost and abandoned objects detection,
• Assessment of the use of facial verification as a token for segregated two steps topology,
• Assessment of interconnection and interoperability between different enrolment and verification schemes (self enrolment+eGate, self enrolment+mobile, enrolment station+eGate, enrolment station+mobile),
• Assessment of ergonomics and human-machine interaction (self enrolment, travel document handling, self biometric capture),
• Assessment of the impact on the current border guard processes and systems.
URJC developed an evaluation framework oriented to the measurement of the system response to antispoofing attacks. This theoretical approach was adapted to ABC particularities, considering the pilot architecture.
The main tasks developed during the pilot were:
▪ Several antispoofing attacks were emulated in the real pilot, studying the influence of several parameters over the presentation attack detection capability of the system. The attack considered and performed in the pilot were:
• Printed face on a paper sheet. Sometimes a printed face is shown with eyes cropped out so that the subject’s eyes blink underneath.
• Digital face displayed on a screen from digital devices such as tablets, smartphones, and laptops. This kind of face presentation can be static or a video. In video attacks facial movements, eye blinking, mouth/lip movements or expressions are simulated.
• A paper 3D mask, emulating the response to the system to a close to real texture with a different 3D shape.
• A high quality 3D mask (silicon) specifically moulded for a targeted face.
• Images attack camouflaged in some part of the clothes, like a shirt.
▪ A systematic observation of the travelers and it interaction with the registration and verification kiosk were performed. The standard model “human-biometric-sensor interaction (HBSI) evaluation method” was adapted to the ABC scenario defining some metrics.
The number and technical level of the presentation attacks considered has been bigger that the initially estimation, and the effort dedicated to this task is slightly bigger than the estimation in the project proposal. This increment of pm doesn’t imply an increment of the global EC funds to URJC because the current URJC pm cost is lower to the initially supposed. The final team isn’t as senior as it was initially expected in the proposal, so the pm cost is lower.
Third period (2017-2018)
WP1 LEGAL, SOCIAL AND ETHICAL ASPECTS
Summary of the tasks carried out and significant results
In the framework for Deliverable 1.9 the following tasks were conducted:
• Updating ABC legal frameworks: data protection and automated border control.
• Adapting ethical assessment methodology to WP5 specifications and requirements. Eticas studied the integration and deployment of the Registered Traveller Programme and Entry/Exit System within the ABC4EU project and the concept for upgraded ABCs according to each type of border and passenger requirements.
• Subsystems analysis: data life cycle reconstruction, examination of performance and identification of vulnerabilities. This assessment focused on the relative data protection compliance of these systems, privacy concerns derived from sociotechnical interaction and their data management protocols when deployed. In this context, particular attention was given to the social impact of biometric processes.
• Constructing recommendations on this basis
In parallel with these assessments, three main groups of tasks were carried out within the package:
- Tasks concerning logistical aspects of the project (including organizing and attending internal and advisory board meetings and the elaboration of reports about the project´s evolution);
- Tasks supporting project partners in tackling ethical aspects of the field research (including the development of consent forms and analysis of the materials provided by other partners with a data protection focus);
- Tasks for dissemination (including publications and attendance to conference and workshops).
Deliverable 1.10 includes assessment of the pilots’ execution as part of the pilot evaluation carried out in WP6, in order to complete the examination of the ABC system implementation and to measure its potential social impact. In this context, Eticas designed and applied a survey and a number of non-participant observations in the various locations (sea border, land border and air border locations) where the system was pre-tested and tested. The methodological approach developed by Eticas for the fieldwork focused on the experiences of the travellers, their opinions and expectations. It was constructed considering a wide set of variables about acceptability, desirability, functionality and adaptability to vulnerable groups. Moreover, the survey was designed to consider the different inputs offered by the project partners and integrated information on variables required by them within the data collection tools.
217 surveys were administered during the final test phase at the Lisbon airport, Algeciras port and Calafat land border. The results of this final research process, as well as the 39 surveys and non-participant observations carried out during the pre-test phase, are reflected in D1.10. Besides this, the document includes an analysis of these findings contrasting them with the information about the system validation provided by the project partner PwC and the results of the entire pilot tests, including the ones conducted in Algeciras. The analysis of all this relevant information was also analysed against previous projects conducting field research in this area (FASTPASS, BODEGA, EU Lisa and others), as well as considering an analysis of the border guards’ opinions on the current transition towards an automated border system. The Deliverable thus constructs preliminary and explanatory correlations about the acceptability of the proposed solution against a selection of sociodemographic variables.
Regarding D1.11 the final report on the social, legal and ethical implications of the ABC4EU project includes a thorough review of our findings throughout the project. It also includes a systematic review of the legal state-of-the-art as it affects the ABC4EU system. This was particularly important considering that the legal framework underwent a significant number of important changes during the project' timeline. An analysis of the labour rights of the border guards and how these might be affected by the implementation of the ABC4EU system was also carried out during this period. Finally, a conceptual framework for interpreting the full range of the social, legal and ethical implications of the ABC4EU system was developed and incorporated into the report.
To conclude, Eticas carried out many tasks concerning the ethics compliance of the project fieldwork. The work of Eticas in ABC4EU has been mainly driven towards building the legal and ethical references to be taken into consideration for its development and towards monitoring the compliance of these principles and regulations in terms of the technological design. The tasks developed in this context included:
a) the design, translation and adaptation of consent forms for all the pre-pilot and pilot tests,
b) the co-design, together with the project partners, of the notification about data management specifications for each pilot test to the Data Protection Authorities in Spain and the monitoring of these notifications for all the countries where the tests were conducted (Spain, Portugal and Romania), and
c) the design and gathering of a non-disclosure agreement to be signed by all the partners involved in the Calafat pilots, which was expressly requested by the Romanian Police during the conducting of fieldwork in this country.
These protocols aimed to guarantee that all partners’ interventions in the development of the pilots was in line with data protection and human rights regulations and principles. The Annex of D1.10 includes a thorough explanation and refers to other contributions made in terms of protocols for human-machine interaction in the system´s implementation, the compliance of fieldwork activities with informed consent protocols, the minimization and deletion of personal data and putting safeguards in place for the treatment of all actors involved.
WP2 TECHNOLOGY ASSESSMENT AND END USERS REQUIREMENT DEFINITION
Summary of the tasks carried out and significant results
The D2.8 is intended to be an EU Border Guard common set of requirements for Strategic and supra-national coordination. Specifically, it aims to provide information about needs for strategic analysis and mechanisms to share such information between the BCP and the national back-office, and between national offices and an EU coordination centre.
These goals were already mentioned for D2.7 but in this case it was the second iteration. Concerning D2.8 the purpose pursued was to enhance and enrich the set of EU border common guard requirements for strategic and supra-national coordination presented in deliverable D2.7 with EU-wide application including all types of BCP, e.g. airports, land and ports.
It is important to highlight that the set requirements were collected by considering regulations, frameworks, international standards and projects as a reference to support them at each level of the final Governance Board structure defined in D2.8.
As second major input, since WP6 activities had to be performed in parallel, PwC decided to take advantage of the information generated during pilots for this deliverable (pilot experience). In this regard, it was possible to obtain information from different sources such as feedback provided by the border guards (CFRONT, CNP and SEF), checklists developed for the first PoC, logs collected by the technical partners from different systems' topologies (the systems' deployment depended on the pilots location), and observation by PwC team.
In addition, regarding WP6 activities, it was gathered all the information obtained from the meetings held with stakeholders and end users, considering the BG approach. In this regard, the meetings conducted were focused on planning the PoC activities.
The deliverable 2.8 meant two major challenges. On the one hand, it was necessary to investigate and obtain accurate and wide information referred to strategic and supra-national coordination. On the other hand, PwC had to perform the WP6 activities in parallel to this deliverable (so it was decided to take advantage of the pilot experience for the D2.8 as mentioned before).
WP5 SUBSYSTEMS DEVELOPMENT & INTEGRATION
Summary of the tasks carried out and significant results
The goals for this period were to achieve the development of final solutions to be deployed in the final Pilots, comprehending:
o Final Biometric Management System
o Final Travel Document Management System
o Final Entry-Exit System
o Final NFP System
o Final ABC System
o Final Mobile System
Within WP5 (Subsystems Development & Integration), research and development activities have been carried out in order to improve the biometric algorithms taking into account the results from the proofs of concept performed at Madrid, Lisbon and Algeciras. These algorithms, which include biometric fusion, facial anti-spoofing, tailgating detection, lost objects detection, and facial recognition on the move for two-steps topology, were fully integrated into the final prototypes. The CES is being extended with an innovative feature that pursues that an e-passport can be trusted even if the necessary master list is not available. The EES has been extended by adding AFIS capabilities and other features. The ABC prototypes have been extended to cover additional use cases (like reading electronic residence permits). The mobile device, the NFP and the Enrolment Station were also reviewed and extended for the final pilots.
Activity 5.1 Biometrics
After Algeciras PoC, Indra agreed with UMIL, URJC, Cognitec, Dermalog and Atos the functionality for the biometric fusion algorithm. UMIL developed two versions of the algorithm in collaboration with Cognitec, Dermalog and URJC. A basic version only takes into account the facial and fingerprint verification scores, and a second, more advanced version, takes as input also the quality of the images. The algorithm for calculating the quality of facial images was designed by Cognitec, while the method for getting the quality for fingerprint images was provided by Dermalog.
UMIL has focused its research in the analysis and development of biometric multimodal fusion methods suitable for ABC scenarios, including the investigation of different fusion strategies, performance evaluation procedures, and early tests ran on biometric data representative of ABC conditions. As a first step, UMIL has studied the suitability of fusion methods at different levels, concluding that score level fusion methods and hybrids of score and feature level methods are the most suitable. Among these, UMIL implemented and tested seven different methods. The methods were chosen paying attention not only to their effectiveness, but also to the possibility of using them in a privacy compliant way. Among the tested methods, UMIL considers as good solutions the application of two final fusion methods detailed in the deliverables. UMIL has also provided illustrative implementations in C language of the selected methods preliminary trained on the biometric match scores computed on an external public dataset using the same biometric matching software used in ABC systems. Preliminary normalization of the scores were also considered. In addition, UMIL has provided suggestions on which thresholds can obtain a good performance, based on the training results. UMIL has also proposed a procedure for increasing the biometric recognition accuracy in ABC systems with hierarchical biometric recognition.
URJC research team has been mainly focused in the improvement of current ABC systems for presentation attack detection in facial biometrics. Different methods to detect spoofing attacks in the case of facial biometrics have been analyzed and an initial proof of concept has been developed that can be modified to be included in the ABC gate. After a deep analysis of the possibilities available, four types of vision sensor (RGB, IR, NIR and depth information) were selected and its application to presentation attack detection were studied. Considering the restrictions and specifications of an ABC gate, it was decided to work with still images.
Two different approaches were researched for the purposes of presentation attack detection, texture-based and geometrical methods. A wide range of attacks with different technical level were considered: printed face on a paper sheet (as baseline method); digital face displayed on a screen from digital devices such as tablets, smartphones, and laptops; 3D masks (paper, silicon, cast, rubber etc) specifically molded for a targeted face.
Apart from participating in the discussion of fusion algorithms, Dermalog carried out the following activities:
• Research and improvements of the fake detection technology.
• Research and improvements of “Aging” technology – fingerprint matching under 12 years.
• Development of a new release of Web AFIS with improvements on the performance and the interface to other systems (e.g. BMS), and including technologies such as the new fake detection technology and the “Aging – Fingerprint matching under 12 years” and.Testing of the new VF1 multi-scanner
The tests done during Barajas pilot allow a valuable experience in order to improve the PAD algorithm and a more advanced approach was carried out by including presentation attack detection on the fly. This new algorithm was integrated in Algeciras final pilot.
The user interaction with the mobile system was deeply analysed during the Romanian pilot, measuring the border guard experience using and adapted methodology based on biometric user interaction.
Indra worked on improving its own algorithms (lost object detection, tailgating detection and facial recognition on the move for the two-steps segregated topology) taking into account the feedback from the PoC. Calibration, configuration and logging tools have been extended so that the algorithms can be fine-tuned easily.
Activity 5.2 Travel Documents Management
The operations needed for Terminal Authentication (GetCertificateChain() and GetTaSignature()) were implemented in the CES in order to prepare it for reading fingerprints from travel documents. The CES server at El Escorial was updated and tests were done with Spanish preproduction passports lent by CNP.
One of the main outcomes of the PoC in Algeciras was the possibility to extend the CES with a functionality that would enable to trust a passport when the CSCA certificate is missing (as is the case for Morocco and some other countries). The ground for this proposal, which was defined in collaboration with Atos (interested in integrating it both in the mobile solution and in the Enrolment Station) was the vast majority of travelers observed during the PoC pilot testing (Moroccan) coming from a country of origin which did not provide any kind of certificate required to validate the authenticity of the electronic and cryptographic material of the passport. The problem is, when this situation happens at a border crossing point, the possibilities of automation are minimal (almost 100% of the travelers in Algeciras are from Morocco). A new operation in the CES (PossiblyValidDS()) has been implemented by Indra. This innovative feature has been evaluated in the final pilots, resulting in a proposal for harmonization that will help automation under this kind of situation. Indra took advantage of this feature, which has been integrated into the Verification Kiosk and Atos did the same in the Enrolment Station (for validating a DS certificate as a candidate to be trusted) and in the Mobile System (for passport verification).
A management console for the security patterns database was developed by Indra, which allows editing the patterns with a user-friendly GUI.
Indra also replaced the current electronic documents inspection system with the CES development in the Indra kiosks. For the final pilots testing and validation, Indra kiosks and Atos Enrolment Station (and the Mobile device, as in the PoC) used the BSI interface provided by the CES, including the PossiblyValidDS() extension.
Activity 5.3 Registered Traveller Programme subsystem development
Atos has continued activity 5.3 focusing primarily on the PoC pilot results and feedback from the BG to improve the Enrolment Station (ES), the user experience and time needed to enroll TCNs in EES and NFP. Therefore, the ES GUI was completely redesigned to have a more intuitive look and feel as well as separating logical functions eMRTD & Visa Reading, EES and NFP into separate tabs instead of the one common application form.
Additionally there were new functions added as outlined below:
• authorization token for access to all backend services controlled by the Front End GW
• added English language capability
• web server functionality improvement
• added details query for TCNs previously enrolled in EES and show their enrolled details.
• added duplicates check in EES and ability to store new eMRTDs in the case that a TCN has an additional eMRTD that can be registered
• added EES crossing history query
• implement new CES service to validate DSC certificate to be saved in a trusted CES DB store for ePassports issued by countries that do not make public their CSCA certificates.
• added NFP details query to show TCNs data previously enrolled
The FE GW was updated to provide a local authorization check to all back end national and central services, so that only BG police previously authenticated against national police systems will be authorized to access a service behind the FE GW. The FE GW was also updated for all new EES, CES & NFP services and updates to existing ones as they were made available by each partner.
The NFP Service was updated for more management functions to be able to read the TCNs file and check if the status is Rejected or Approved as both are supported in the Final Pilot NFP lifecycle.
• EES & NFP Enrolment Station (ES) with its complete re-design to have a more intuitive look and feel and separation of logical functions eMRTD & Visa Reading, EES and NFP into separate tabs instead of the one common application form. All new functionality and service queries have finished implementation as described previously in the previous interim report.
• FE GW updated to implement all new and updated services towards the EES, CES & NFP subsystems, and an authorization token for controlling access to all national and central EU services over the FE GW.
• NFP Service updated for new service to read the TCNs file and NFP lifecycle management for rejected applicants.
Activity 5.4. Entry/Exit subsystem development.
Activity goals focused on addressing the PoC pilot execution results and including also additional features that were not included in the first version.
The main findings from the Proof of Concept trial execution was that fingers collection process needs to accommodate cases where the slap capture is not possible, namely
• People with amputated fingers or with medical conditions that require that the system allows them to be enrolled even without providing fingerprints.
• People with temporarily medical conditions or immediate difficulties that do not allow them to enrol their fingerprints need to be allowed to register even without providing their fingerprints with the condition that they will have to provide them on the next Entry.
The final solution also had additional features that were not implemented and tested in the POC Pilots, namely:
• Duplicate search on existing accounts:
o The final solution that was developed, deployed and piloted in the final trials is person centric, meaning that each Passenger has unique account to which multiple Passports and Visas can be associated. This solution evolved from the document centric approach, in which the same Passenger with multiple documents could create different EES accounts, each associated to each of the documents.
• Enrolment Process Supervision
o A monitoring station was developed in order to allow the enrolment process supervision by the Border Guards. This monitoring station aimed not only to follow the Passenger’s enrolment process results but also to allow the Border Guard to intervene in the enrolment process and support the Passenger’s in need.
▪ For the cases where the Passenger was not able to provide fingerprints, the Monitoring station allowed the supervising officer to exempt them to provide fingerprints. This exemption could temporary or permanent, depending on the Passengers case.
• Entry and Exit Crossing:
o The pilot of EES final solution was deployed both Arrivals and Departures, meaning that both Entry and Exit records were registered, which brought the opportunity to develop, deploy and test new features, namely:
▪ Verification of Entries without Exit - In the event of a new Entry of a pre-registered Passenger, EES assesses if there were any previous entry occurrences that do not have an Exit Record associated to it.
▪ Passenger Border Crossing History – EES system was also able to present more information to the Border Guard, namely the history of previous transactions, to help assisting on the Border Control Process decision.
Activity 5.5. Existing ABC upgrades (Airport, land, sea).
Activity goals focused on addressing the PoC pilot execution results and preparing the ABC systems to be in compliance with the adaptations required for EES Final Prototype.
The use case considered for this Final Solution was based on the harmonized automated border control processing of Third Country Nationals Visa Exempt and Third Country National Visa Holders.
The objectives for this activity were the following
▪ Ensure that the Updated ABC System Final Prototype developed during the project complied with all technical and operational requirements.
▪ Test the system performance.
▪ Provide an overview of how the updated ABC System would operate in a pre-operational environment.
▪ Identify possible limitations, risks and exceptions in the mobile inspector.
▪ Provide testing results to set of recommendations to be used as the basis for an E.U. harmonised/standard design and usage of mobile systems.
Indra worked on integrating new hardware prototypes:
• New automated doors. After asking different gate vendors for changes in their communication protocols in order to accommodate to Indra’s requirements, different doors were integrated and tested in order to harmonize the interfaces for the integration of the gates with the software.
• New facial on-the-fly verification module. This piece of hardware, which is in charge of facial recognition, has been designed in a way that will allow to be used in different configurations. A preliminary prototype was built and tested integrated in a gate. Indra worked on a second prototype, with the form factor of a kiosk, which was tested in the final pilots in a segregated two steps topology. The software for controlling the new facial verification module was also developed.
Indra was working on several aspects that were not covered in the PoC:
• Integration of tailgating detection algorithms. Although preliminary versions of this algorithm were tested at the PoC, they were not integrated with the ABC software during the first pilots.
• Integration of biometric fusion algorithms developed by UMIL. The fingerprint will be asked only if the facial recognition was both not good enough and not too bad.
• Integration of anti-spoofing (presentation attack) algorithm developed by URJC. This presentation attack algorithm has been tested in Algeciras pilots, taking the Calafat pilots as the baseline for the comparison of the results.
• Reading visa stickers at the EES enrolment kiosk.
• Integrating new EES functionality, like performing an AFIS search before enrolling a new document for avoiding duplicates (passport renewal, multiple nationalities, etc).
• The integration of the new Dermalog VF1 two-in-one (eMRTD + fingerprint) scanner has been evaluated, but not accomplished. The intention was to demonstrate the integration of a state of the art equipment, which could greatly facilitate the enrolment of third country nationals, while reducing the size of the systems at the same time. However, the device was not mature enough and its integration was withdrawn during the period.
Activity 5.6 Mobile subsystem development
Atos commenced activity 5.6 by reviewing thoroughly the BG feedback and resultantly have re-designed the GUI to much improve the BG experience and make items of attention (e.g. alerts) more obvious while reducing clicks and improving processing times by querying external DBs in parallel, and thus sending queries as soon as the required data is available rather than having a more process flow approach as implemented in the first PoC. To facilitate this, the GUI design will now make more use of tabs to enable this user experience, similar to what was observed in the Mobile Pass application. In the Final Pilot application, the BG decisions are now only needed upon initial reading of the eMRTD (so to continue the process or to end it and refer the traveller to the manual booth or enrolment or refer to 2nd line), and at the final step to allow or refuse the traveller to cross the border or refer to second line. Further to BG and traveller user experience, improvements have been made also in presenting a graphical view of fingers stored in EES and VIS available for checking against.
Additionally there are new functions added as outlined below:
• Offline mode and synchronisation of Offline Traveller Records
• New EES Exit workflow
• Biometric Fusion
• eMRTD reading extended to eRP and eID cards (ICAO compliant)
• New CES responses for online Passive AuthN indicating CSCA not available but the DSC is found in Trusted DSC store
• Face matching service (Cognitect FaceVACS)
• New Authorisation Token for access to all backend services & DBs
• New 3rd party SDKs tested:
o New Imense Camera OCR reading of Visas and Number Plates
• Updated FE GW Interfaces:
o Authorisation token to back end service & DBs
o Certificate Exchange System for Online Passive Authentication (BSI TR-03129-2 )
o Entry Exit Service (EES) updated and new services incl. EES Crossing History, 1:1 FaceMatch
o NFP Service updated for new services
Activity 5.7. RTP, EES and upgraded ABCs integration.
Activity goals focused on addressing the PoC pilot and final pilot execution results and preparing the integration between ABC4EU Touchpoint solutions (ABC systems, Mobile Stations and Enrolment Stations) to be in compliance with the new features that were not included in the first version of EES, BMS NFP and TDMS/CES subsystems.
The integration activities ended with the preparation/deployment of the Pilot Solutions (touchpoint and backend systems installation).
With the change of EES to a traveller account based system rather than a travel document based system this meant a complete set of new EES Services that had to be integrated and debugged through testing to the Front End Gateway (FEGW) and the Enrolment Station and Mobile System (end clients). Additionally new services were made available for biometric face matching in EES/BMS for both TCN VE & VHs as part of the biometric fusion capability integrated on the Mobile System, and biometric duplicate search on the Enrolment Station integrated with EES/BMS.
WP6 BCP Test & Validation
Summary of the tasks carried out and significant results
PwC effort was focused on continue with the validation activities for the second round of pilots and the elaboration of related reports.
The checklists of the validation activities that were elaborated internally by PwC (based on the information included in deliverables 6.2 6.3 and 6.4) were updated once the results collected from Madrid and Lisbon (first round) pilots were analysed.
Likewise, PwC updated the set of forms elaborated to facilitate the collection of data splitting the uses cases in smaller tests. For this purpose, in addition to the forms elaborated, the observation was a key method to collect these results properly along the first PoC.
The validation results and the tests performed for the systems involved in the first PoC include:
- User and ergonomic tests
- Functional tests
- Mobile integrated solution tests
- System security tests
- Data protection tests
In addition, PwC was responsible for analysing the feedback collected by Eticas from the travellers that agreed to participate along the first PoC. Likewise, PwC gathered the border guards’ feedback from the different locations (Madrid, Lisbon and Algeciras).
Potential Impact:
Potential impact
It has been identified since the preparation of the proposal that the change in the regulation regarding border management, and the development and integration of a central, EU-wide, Entry Exit System, and a national facilitation program, will imply a huge change in border activities in terms of human resources, time required for border clearance, annoyance to the travellers (as additional processes have to be carried out, prior or during the border checks) and footprint for new equipment at the border crossing points, among others.
One of the main objectives of the project was to assess this impact by the deployment of functional systems covering the technical scope of the proposed solutions, under real operational scenarios, in order to identify further ways of automation that facilitate and alleviate the negative impact of its deployment.
As it has been observed during conversations and surveys with the travellers during the pilot execution, extending the automation to third country nationals is very welcome by the travellers, as the usually feel that TCNs are discriminated in some way against EU citizens because they can’t avoid waiting at the border queues.
Systems such as those developed in ABC4EU project, deployed at the borders integrated with the Entry Exit System and National Facilitation Program will help ensuring a seamless border crossing for third country nationals while keeping the security levels at the borders.
Overview of dissemination activities and exploitation of results
Overview of main dissemination activities & channels.
In accordance with the DoW, the following major dissemination activities were performed within the entire project’s duration and in part they will continue after the project has ended:
Official Logo and Official ABC4EU project website.
Dissemination Materials, such as hard copies of Leaflets, posters, flyers and roll-ups created and distributed in various events such as pilots, conferences, workshops and so forth.
Marketing Material such as: Shoe bags, luggage tags, and small bottles for storing liquids for travellers, all bearing the ABC4EU project logo, as a way to promote the project have been distributed in conferences, abc4eu final conference, end user workshops and project pilots.
Social media promotions of project milestones, events, or other information worth sharing with the public audience through project Website, Facebook, Twitter, LinkedIn and YouTube.
Forum
Forum has been created in order to promote discussions from a wider audience regarding privacy, technology used and legal aspects of ABC gates. Link to the forum: www.forum.abc4eu.com However due to lack of interest it has been discontinued half way through the project.
Twitter
Twitter is a social media platform that is used to quickly share short status or news reports and/or refer traffic to the main project website. Twitter has been chosen for its short and quick communication abilities without giving too much information away. Also another reason is that twitter is used actively by most of our partners and end users.
Twitter name: @ABC4EU
Facebook
Facebook is a social media that has been primarily used to share public deliverables and make any announcements necessary related to the project. For example newsletters, press releases, scientific publications, presentations and other dissemination activities performed by ABC4EU partners have been shared there. The project link is: https://www.facebook.com/abc4eu
LinkedIn
LinkedIn is a professional networking platform. Laurea has created an ABC4EU-project group to which all news has been shared. LinkedIn has been used to promote the seminar to expand the end-user community. Also, external links, photos and such material. LinkedIn name: ABC4EU-project
YouTube channel:
Several videos showcasing the ABC4EU project technology, process flow, and other project related information have been produced and uploaded to social media and project YouTube channel. These have also been published in the project website.
.
Newsletters: Considering the newsletter definition, it is a regularly distributed publication about a particular topic which is of specific interest to receivers. Thus in context of ABC4EU project, thematic newsletter has been distributed regularly in order to notify relevant parties regarding the project progress. These were also published in project website, social media and sent to project subscribers.
Press releases: relevant media channels have been identified and then selected according to their best suitability to the project needs. In the case of ABC4EU project all varieties of published short releases have been named “press release” independent of the media type. At least seven press releases have been released during the project.
Events: The ABC4EU project has been effectively presented within relevant conferences, seminars, workshops, fairs which have been thematically related to the project’s scope. The main goal of this participation was to raise awareness of the project among different target groups to the best of abilities. Within such events ABC4EU project has been promoted. The presentations have been concentrated on the area of interest of event participants. The project has been present both in research conferences and national events addressed to practitioners I.e. Students, public administration employees, private enterprise employees, academic institutions representatives and so forth. Events such as keynote speeches, paper publications, conferences, presentations, lectures and fairs, have been conducted periodically whenever possible such as the SRIEE 2017 event and the Nordic Business forum.
Pilots such as dissemination of the project via direct contact with various end users such as airport operators, border guards and travellers. In These events all project promotion materials have been distributed such as leaflets, flyers, roll-ups especially created for the pilots, shoe bags, liquid bottles and luggage tags.
Overview of main exploitation activities
The main exploitable outputs have been conducted via
Development and improvement of dissemination strategies, through lessons learned methodology for better future dissemination outputs.
Technology development whose applicability is not only limited to ABC gates but to a wider usage area, such as algorithms for abandoned objects for increased security, look&walk for two step segregated topology and so forth.
ABC4EU consortium have promoted the project results through papers, articles and publications on specialised newspapers and journals.
The objective has been on one hand to obtain a great visibility among research communities and target groups (avoiding wasting resources disseminating the project through non-specialised press communications), and on the other hand, it has been to inform and reach research communities and experts in the targeted sectors through scientific journals. All published articles are reported in the website section “publications”. Also after the project has ended statistical information from traveller interviews and workshops will be used for publishing new papers regarding ABC gates and the EES.
Skills development, training for new staff at airports, mobile system development SDK training for employees, Teaching I.e. integration of the project with the study units at universities, incorporation of the acquired knowledge in educational activities, such as BA, MS and PhD programs and summer schools.
Learning from experience such as: Applying lessons learned methodology and carrying out detailed analysis in order to take advantage of the conclusions and results obtained in order to address the latest tests planned in the project in the adequate manner.
Contacts with other FP7 / Horizon2020 –projects
Project partners created and maintained contacts with other FP 7- projects in order to share experiences. Joint activities included but were not limited to:
• Cross-referencing other projects’ websites. In this way the visitors of a project website could be easily directed to other similar projects funded by FP 7 and find out about projects they were not previously aware of.
• Distributing promotional material to other projects events. Through such cooperation, distribution of leaflets and promotion of the project is easily multiplied with minimum effort and without any additional costs. Moreover, it provides the opportunity to disseminate to events that none of the partners is present and to countries outside the consortium.
• Joint participation at seminars and conferences.
Cooperation with other similar projects together with end users from academia, legislation, border security and ethical experts, in order to share information, learn from one another and discuss project results and expectations from end users, legislation, ethics and technological perspectives.
Dissemination materials
Project Logo
The project logo was developed in the beginning of the project. The general shape was agreed by the project partners that the logo shall refer to the main issue of the project. The main goal of the project logo is to create awareness of the ABC4EU and attract the attention of potential target groups. The logo is clear and simple and may be easily recognised and relevantly associated. It contains the abbreviation of the project which includes ABC4EU name and the e-Passport Chip.
The logo has been provided in different formats and resolutions to be comfortably suitable for different purposes i.e. document, website, print and so on.
Website
The ABC4EU website (www.abc4eu.com) has been the main part of the project dissemination. The website consists of two areas: the public and private. From the dissemination’s perspective the most important part is the public side of the website which is available for the general public and will contain information related to the project e.g. a project summary, news, information on the consortium, public publications and so forth.
The project website is mainly utilized to raise the awareness among target groups’ representatives as well as to be a source of information related to events, publications and so on.
There are three main goals for the website:
• Ability to provide detailed information concerning the project aims, progress and news
• Hold publications in disposal for any interested party
• Collect internal documents and ensure their availability among the project partners.
The website has been regularly updated, news has been introduced, publications have been published, and the project’s progress has been generally reported.
Poster
The poster has been adjusted to the current project’s stage and has been promoting current project’s achievements, it has also been adjusted to the target audience in case of preparations for a previously planned larger event.
Poster’s aim is:
• To inform on current project’s stage and achievements;
• To attract stakeholders and draw their attention to the ABC4EU project, its goals, outcomes and achievements;
• To share only the key issues of the project and to encourage people in finding more information (e.g. visit project's website, attend the course or any event where the project will be presented.
The first project’s poster, prepared at the very beginning phase of the project was concentrated on the general project’s purpose. Its aim was to announce the project, its goals and planned outcomes. It is addressed to the general public and to researchers.
The poster schema is consistent with the project logo and leaflet colours and character. The picture represents the initial poster’s appearance and content.
Roll-Up
Laurea UAS has also developed roll-ups according to the uniform design of all ABC4EU dissemination material. This material may also be printed by partners and used in events where ABC4EU is presented. The Roll-up has been created with the aim of attracting potential interested parties in events such as fairs, project presentations, lobbies on meeting venues, and also have been used in pilots for providing more information to travellers as well as to attract their attention.
The first roll-up has been created on M9 as an addition to the overall dissemination material. This type of material shall include the project logo, marketing phrases and funder information.
Leaflet
The ABC4EU leaflet has been adjusted several times to the current project’s stage and promotes current project’s achievements. Throughout the project there has been also possibly to adjust the leaflet to the target audience in case of preparations for a previously planned larger event.
The first project leaflet has been prepared at the very beginning phase of the project and is concentrated on the general purpose of the project. Its aim is to announce the project, its goals and planned outcomes. It is addressed to the general public and to researchers.
Towards the end of the project, particular outputs have been described in the leaflet, in a timely manner.
Flyer
A flyer has also been added to the overall dissemination material by the M9. Similar to the roll-up, the flyer is designed as a quick touch point to the project. It consists minimal information about the project including: project logo, funding, marketing phrase as well as addresses to the online presence of the project. The leaflet has been designed to attract the attention of readers and increase their interest towards receiving more information regarding the project.
Marketing Materials
Marketing materials have been shared in various events where ABC4EU project has participated, especially in the pilots, where materials were given to travellers after they had accepted
List of Websites:
www.abc4eu.eu
Twitter name: @ABC4EU
Facebook: https://www.facebook.com/abc4eu
ABC4EU project started in January 2014 to address the topic SEC-2012.3.4-6 Enhancing the workflow and functionalities of Automated Border Control (ABC) gates, in the context of the 7th Framework Programme.
Before the kick-off of the project, many ABC gates systems were deployed in the main European airports, not only as pilot projects intended to test the capability of these systems to improve the border crossing processes in aspects such as speed, security, process automation, false rejection reduction, etc.; but also as fully operational deployments, designed to facilitate the border crossing process to bona-fide travellers while maintaining the security at the border.
One of the issues to be addressed by ABC4EU project has been the need for a harmonised approach in areas as e-passports management, biometrics, gate design, human interface, processes, PKD certificate exchange, signalling and interoperability.
On the other hand, the EC published in 2011 a new approach for border management which implied an impact in ABC systems development (Smart Borders initiative, formerly known as Stronger and Smarter Borders), including an integrated, EU-wide EES mainly aimed at extending the use of automated systems to third country nationals, while helping identifying irregular immigration, apprehending overstayers, and generating strategic information to fight and prevent criminal activities. In the scope of the initiative, a Registered Traveller Program (formerly known as National Facilitation Program) was conceived to enhance automation of the border checks, aimed at frequent, pre-screened and pre-vetted third country nationals crossing the borders.
These topics were addressed in the different stages in the project:
• Assessing the impact on the privacy and legislation of the project activities
• Analysing the state of the art at the technical level, as well as lessons learnt of previous pilot and operational projects deployed looking for further automation
• Designing the evolution of the systems to be adapted to the changing legislation
• Setting up the overall architecture of the subsystems developed
• Adapting the systems to achieve the goals of the project
• Executing pilots to assess the technical developments in two stages, one for testing prototypes and the second for final systems, in operational conditions
• Ensuring proper dissemination in workshops, seminars, etc of the activities and outcomes of the project
The impact of the project development at the privacy and personal data management dimensions was early anticipated and addressed since the very beginning of the project via a transversal activity, by defining a framework for privacy management, training and continuous overview of the activities of the project and assessment of the outcomes of the partners’ activities during the project execution.
Project Context and Objectives:
The project objectives were in line with the topics and issues to be addressed in the call SEC-2012.3.4-6 Enhancing the workflow and functionalities of Automated Border Control (ABC) gates:
• The development of a Smart Borders initiative includes establishing a central Entry-Exit System and a Registered Travellers Programme which facilitates border crossings of TCNs.
• The ABC Systems currently deployed at some of the external borders regularly uses only one biometric identifier (face or fingerprints)
• Can only be used by EU citizens holding e-Passports or National ID cards.
• The future EU policy will require the processing of TCNs at the external borders by verifying their biometrics through automated processing (ABC eGates).
• UE citizens will also be able to make use of second generation e-Passports including multiple biometric identifiers.
• ABC systems implemented by some Member States allow for automated border checks based only on facial image verification.
• Moreover, the user interface is not being harmonised among Member States.
• This pilot should address these two issues: to propose a harmonised and coherent solution, allowing using the e-Passports to their fullest potential during the automated processing.
• Attention should be given to compliance with European societal values, citizen’s rights including privacy and acceptability.
ABC4EU project consortium was originally composed by 15 partners from the industry, academia and end-users from 8 EU countries, and changed during the project execution. Two partners left the project during the first year, one industry partner joined to take over certain activities and a fifth end-user engaged with the project at the later stage for supporting diversification of border crossing points and countries where the pilots were deployed and validated.
The main goals of the project were:
• Analyse the ABC concept vs EU legislation, in particular the Schengen Border Code, to identify which processes can / cannot be automated and the impact that different levels of automation could have in the SBC procedures.
• Optimize and harmonize the ABC processes related to the increasing flow of people crossing borders, without undue delay and with minimal intrusion.
• Assess the feasibility of a future National Facilitation Program (NFP) and the Entry / Exit System (EES) in the EU which could be integrated into the ABC gates from technical, functional and ethical point of view.
• Evaluate, optimize and harmonize the use of 2nd generation passports biometrics (facial and fingerprint) for passengers identity verification.
• Identification and design of a basic set of features to achieve a common user experience, involving ABC infrastructure, passenger verification processes, interface look and feel, signalling, etc.
• Identification and, to the extent possible, overcoming of obstacles which are preventing today the certification distribution throughout Europe to access the fingerprint information in 2nd generation ePassports.
• The compliance of the proposed processes with the legal requirements such as respect for fundamental rights, in particular the protection of personal data and with the ethical principles.
• Design of a C4I concept for regulated borders, integrating EES, NFP and ABC gates.
ABC Gates
The activities in the scope of ABC Gates redesign were mainly defining a harmonized approach in areas as e-Passport management, biometrics, gate design, human interface, processes, PKI certificate exchange, signalling and interoperability.
Ensuring the integration of multiple biometrics in the identity verification process, as included in 2nd Gen e-Passports, and to allow third country travellers registration in the EES and NFP.
Anti-spoofing developments, focusing on both biometric identifiers, face and fingerprints, which will allow for enhanced security.
The development of new support systems in line with the Stronger and Smarter borders regulation, fully integrated in the ABC Gate concept.
2nd generation passport
Setting up the technical features to maximize the use of 2nd generation e-Passports to their fullest potential.
Trying to overcome the obstacles that prevent the certificate distribution and efficient travel document inspection systems.
Implementation of multimodal biometric fusion during the identity verification process, while the convenience for the traveller is increased, as different biometric markers are available.
Next generation biometrics
Research on new anti-spoofing techniques, for both fingerprint and facial biometric markers.
Development of multimodal biometric fusion algorithms to efficiently address identity verification of the traveller, aimed at both stationary, regular ABC deployments and mobile devices.
Balancing security/convenience by defining:
• A “passenger experience” oriented scenario with face as main biometry and fingerprint as a fall-back verification when applicable, with faster clearance process cycles.
• A more stricter multimodal fusion scenario for reinforced security, with lengthier clearance process cycles.
Stronger and Smarter borders integration
ABC4EU has developed a National Facilitation Program system, in line with the latest trends of Smart Borders package, aimed at facilitating the border crossing of frequent, pre-screened and pre-vetted third country nationals.
Many of the entry – exit records for the third country nationals have been automatically registered in the ABC Gates, therefore, ABC4EU proposal assumes that the analysis of the practical feasibility of this initiative is been also included in this project.
EES and NFP prototypes have been developed and integrated with several ABC gates in different countries in order to test and validate its feasibility from the legal, functional, technical and ethical perspectives.
Mobile solution prototype
Different kind of mobile devices available in the market, aimed at different user segments, were analyzed to develop an efficient solution for border checks in mobile scenarios.
The aim of the project has been to develop a prototype, handheld-based solution, possible to be used inside trains and buses, while the Border Guard is in motion, covering all border scenarios.
Pilots
The sites for testing have been Portugal (airport), Spain (airport and seaport) and Romania (land border). The aspects of the project that were validated:
• Ergonomics for the operational End User and traveler.
• Document verification (physical and logical) capabilities.
• Biometric identity verification (per type of biometric marker and verification platform/hardware).
• Integration with external systems (certificate exchange, VIS, SIS, EES, NFP).
• Mobile/integrated solutions.
• Central control and support systems.
The pilots were executed in two stages, with the objective of testing a first prototype to be redesigned and reengineered into a final system to be tested in the second pilot round.
Engineering and societal impact
The impact on the ethical and privacy dimensions was early anticipated, even before the project started. In order to address these topics, dedicated ethical and societal review partners established the early inclusion of legal framework at the beginning of the project, including targeted training for designers/engineers to ensure privacy protection and societal awareness by design and ongoing review of every work package by the ethical/societal partners.
The project partners implemented a holistic approach to societal impact, including privacy and data protection but also factors relating to social acceptability and ethics to ensure the success of the project.
Project Results:
First period (2014-2015)
During the first period of the project (from month 1 to month 12) the activities were focused on:
1. Legal, Social and Ethical aspects
After an initial assessment of the procedures for Privacy Management, providing a specific set of guidelines for ABC4EU project, WP1 started with the review of the Legal Frameworks, the review of case studies, providing a comprehensive analysis of ABC/EES/RTP systems deployed throughout Europe with a view on some significant cases outside Europe, the anticipation of externalities and social costs of implementing ABC gates and the analysis on the acceptability and feasibility from legal, ethical and social perspectives. By completing these activities, the contours of the legal, social and ethical assessment of the project began to be laid out. Then, the methodology that developers can use when designing and engineering ABC gates was produced and tested for the first time in the assessment of the work carried out in the scope of ABC Processes and Workflow design.
2. Technology Assessment and End Users Requirements definition
The main objective of the WP2 is to provide the Consortium with a set of common functional requirements for ABC systems from the End User perspective. To achieve this objective, the WP2 defines a series of activities which are aimed to:
• Get the state of the art of technologies which are potentially applicable to Border Control.
• Identify mandatory and forbidden checks to assure the specific functional needs and requirements of End Users at BCPs are performed according to law, covering all types of BCPs.
• Determine the specific functional and information needs and functional requirements at the strategic level for Border Guards, for national and supra-national coordination centres.
• Understand the needs and requirements set out by pan-European initiatives and systems, including existing (SIS/VIS and API) and prospective ones, such as the EES and the RTP.
• Obtain the experience stemming from those Border Guards who already operate ABC systems and biometric-based border controls.
• Understand automation possibilities through an analysis of automation VS security trade-off for the different types of travellers, including those already defined in the Schengen acquis (EU/EEA/CH citizens and Third Country Nationals) and those which would be defined as part of new EU initiatives, such as an RTP.
3. ABC Processes and Workflow design
The global objective of this Work Package is driven by the need of harmonization of ABC systems, in terms of traveller experience, speed up traveller flows and optimize solutions of ABC e-Gates. To achieve this goal, the activities were focused on:
• Define a common architecture for the design and operation of ABCs (e.g. the configuration, operational rules, interface with the travellers, and intervention of the border guard personnel).
• Assess the technology evolution of the biometric data in the passports, including an analysis to the applicability of new features of 1st and 2nd generation of passports.
• Evaluate and assess the usage optimization of the 2nd generation of EU eID cards and residence permits, as well as, the Registered Traveller Token scenarios in order to provide recommendations for Third Country National Unique Identifier in terms of performance, usability and economic viability
• Evaluate and determine the interoperability requirements of a BCP with external subsystems, with a concern in harmonization while designing the corresponding interfaces.
WP1 LEGAL, SOCIAL AND ETHICAL ASPECTS
Objectives for the period
Within WP1, the theoretical and practical elements of the project’s societal impact assessment were carried out:
• It was provided a legal mapping of the EU’s border management and privacy & data protection frameworks, as a means providing project members an understanding of the regulatory framework of ABC.
• Societal impact questions were anticipated by reviewing previous cases of ABC in the EU and beyond, with emphasis also placed on entry-exit systems (EES) and registered traveller programmes (RTP) that might provide best practices for the EU’s planned harmonized implementation of these systems.
• It was provided background on the stakes of border management from the standpoint of questions relating to acceptability, desirability, ethics and data management.
• It was provided a methodology for making other project members aware of societal impact, and provided a practical basis for training relating to legal and ethical questions.
Summary of the tasks carried out and significant results
Activity 1.1 Review of Legal Frameworks regarding technologies and border management
This activity has carried out a comprehensive analysis of the relevant legal framework at both EU and Member States (selected case studies) levels.
Activity 1.2 Review of Case Studies
The aim of this activity was to study the earlier experiences with automated border management so as to lay out the previous challenges faced by those systems, and to anticipate the potential legal, ethical and social externalities of the mismanagement of specific cases.
Activity 1.3 Acceptability and Feasibility from legal, ethical and social perspectives
This activity reviews relevant literature as well as lessons and suggestion from different civil society groups and bodies related to technologies applicable in the scope of ABC4EU in order to start working on a methodology to address ethical and social concerns in the project.
Activity 1.4 Anticipation of externalities and social costs of implementing ABC Gates
This activity explored the interaction of guaranteed rights and device functionality and lay out impact maps at the legal, ethical and social levels, taking a further step in the shaping of the methodology.
Activity 1.5 Methodology
This activity brought together the knowledge produced by the previous tasks in order to produce the methodology that developers can use when designing and engineering ABC gates.
WP2 TECHNOLOGY ASSESSMENT AND END USERS REQUIREMENT DEFINITION
Objectives for the period
The objective is to provide the Consortium with a set of common functional requirements for ABC systems from the End User perspective, not only providing an adequate framework for following WPs, but also serving as a baseline for harmonization and standardization.
Summary of the tasks carried out and significant results
Activity 2.1 Technology Assessment
This activity is aimed to be a survey of the state of the art in the existing and emerging technologies which may have an impact on Automated Border Control systems. Furthermore, the results obtained at the end of this survey are intended not only to lay down the foundations for future decisions about which technologies should be taken into account in the next stages of the ABC4EU project, but also to provide information to determine the actual feasibility of End Users’ requirements. This activity culminated with the main result of the deliverable 2.1 “Technology state of the art survey.”
Activity 2.2 Impact of EU initiatives
This activity is aimed to evaluate the impact of the integration of current EU initiatives in the Border Control Point process and aid to determinate the End-Users requirements according to current European legislation. This activity culminated with the main result of the deliverable 2.2 “Impact of EU Initiatives.”
Activity 2.3 ABC experience. Lessons learnt.
The activity reports on the lessons learnt and best practices in automated border gates obtained from Border Guards and from initiatives from EU, in particular Frontex, and other relevant entities, for example IATA. This activity culminated with the main result of the deliverable 2.3 “ABC experience. Lessons learnt”.
Activity 2.4 Border automation analysis
It is aimed to analyse the activity of crossing a borderland through a Border Control Point (BCP) from a point of view of what sub-process within BCP process are susceptible or are non-susceptible of automation. This activity culminated with the main result of the deliverable 2.4 “BCP process. Automation Analysis”.
Activity 2.5 Study of traveller flows in BCPs (airport, land and ports)
The purpose of this activity is to study traveller flows in different type of Border Crossing Points (BCP) to define guidelines for the harmonization of automated. This activity culminated with the main result of the deliverable 2.5 “Study of traveller flows in BCPs (Airport, land and ports.)”
Activity 2.6 EU common basic set of functional requirements
This activity is aimed to collate and review a common set of requirements with EU-wide applicability, focused on the functional needs of the operational End User at the BCP (be it air, land or sea). This activity culminated with the main result of the deliverable 2.6 “EU common basic set of functional requirements.”
WP3 ABC PROCESSES & WORKFLOW DESIGN
Objectives for the period
The global objective of the WP3 is driven by the need of harmonization of ABC systems, in terms of traveller experience, speed up traveller flows and optimize solutions of ABC eGates. The activities provide to the Consortium a set of processes, workflows and recommendations for the ABC design from the theoretical and practical Consortium partners’ experience, serving as a consolidation of the available ABC knowledge and a baseline for the subsequent work (analysis, design and development WPs).
Summary of the tasks carried out and significant results
Activity 3.1 Harmonized rules for ABC design: Processes, Human interface, signalling
This activity is aimed to define a common architecture for the design and operation of ABCs: the configuration, operational rules, and interface with the travellers and intervention of the border guard personnel. This activity encompasses the following tasks:
• Gathering and description of a common set of business rules which could be applied through all the Border Guard Authorities in order to define a harmonized logic.
• Description of the available signage at airports and during ABC usage (e.g. media options, human factor, usability) in order to reach at a set of recommendations and best practices.
• Assessment of the already implemented ABC designs (e.g. single-step vs integrated two-step vs segregated two-step process, single door vs mantrap configuration, swing doors vs sliding doors).
• Assessment of the specific characteristics of travellers with special needs (e.g. travellers with reduced mobility, visually impaired travellers, minors) and their relationship with the ABC
• Assessment of the streamlining of the human interaction (e.g. travellers with the ABC, border officers with the ABC, travellers with border officers).
• Evaluation of the ABC design with the objective of optimization for facial recognition.
This activity culminated with the main result of the deliverable 3.1 “Guidelines for technological harmonization of ABC: view of the end-user”.
Activity 3.2 Biometrics: Evaluation and optimization of 2nd Gen ePassports
The aim of this activity is to assess the technology evolution of the biometric data in the passports, including an analysis to the applicability of new features of 1st and 2nd generation of passports. This activity encompasses the following tasks:
• Assessment of the quality and efficiency of the passports, specifically the face image quality for facial recognition, fingerprint image quality for fingerprint recognition, passport performance (e.g. security features, average times, electronic authentication) and passport improvements.
• Description of a multi-modal biometrics model and assessment of its advantages, performance, usability and convenience for biometric verification.
• Assessment of biometric passport quality from different countries, through the development of a Gap analysis based on procedures and biometric data specifications (survey analysis) in order to define a set of recommendations to optimize facial and fingerprint recognition.
• Analysis of both ABC and passport working together in order to increase convenience and performance for the traveler, focused on electronic document reader, face image and fingerprint image improvements.
• Assessment of the traveller data protection theme.
This activity culminated with the main result of the deliverable 3.2 “Assessment of the compliance of new biometric data supports for ABC - 2nd Gen ePassports”.
Activity 3.3 Biometrics: Evaluation and optimization of 2nd Gen EU eID cards and residence permits
This activity is aimed to assess the usage optimization of the 2nd generation of EU eID cards and Residence Permits, as well as, the Registered Traveler Token scenarios. This activity encompasses the following tasks:
• Study and analyse the compatibility of eID cards and Residence Permits with ABC, through a Gap analysis and assessment of the different document specifications, in order to provide conclusions and a set of recommendations.
• Evaluation of the face and fingerprint image quality stored in both documents.
• Assessment of the different Registered Traveler Token scenarios, through the usage of three different elements to compare them: performance, usability (from a traveler’s and border guard’s perspective) and economic viability.
• Description of procedures in order to define a set of recommendations to optimize facial and fingerprint recognition.
• Assessment of the societal impact context for the deployment of eID cards, residence permits and the proposed use of Registered Traveller (RT) tokens in the Smart Borders package.
This activity culminated with the main result of the deliverable 3.3 “Assessment of the compliance of new biometric data supports for ABC - 2nd Gen EU eID cards”.
Activity 3.4 Interoperability of ABC points with external systems
The activity 3.4 is aimed to evaluate and determine the interoperability of ABC points with external subsystems, with a concern in harmonization while designing the corresponding interfaces. This activity encompasses the following tasks:
• Description of the different systems that compose an integrated solution for BCPs within the Schengen area and assessment of the interoperability requirements inherent to a BCP, considering both existing and new subsystems.
• Assessment of the horizontal integration of ABCs with border management systems.
• Discuss the technological aspects that should be taken in consideration when implementing the solution: interface technology, software architecture, communication flow and data security.
This activity culminated with the main result of the deliverable 3.4 “Interoperability assessment including C4I”.
WP6 BCP Test & Validation
Objectives for the period
The design and development of a validation framework has been the main goal.
Summary of the tasks carried out and significant results
A research was carried out on this stage with regard to the technical, legal and operational levels, with the final aim to find a set of common criteria, and in this way develop a validation framework of all activities involved in the previous WPs. The structure is in line with relevant governance frameworks and standards in order to provide guidance to top management in setting an automated border control system from a governance perspective.
This deliverable is composed of different parts that involve all data required. The preliminary design and specifications of every step in the validation framework have been described, focusing in a high level structure to define the framework. The report will show the results of the task carried out previously, and define the addressee and the regularity of the reports.
Second period (2015-2016)
WP1 LEGAL, SOCIAL AND ETHICAL ASPECTS
The main objectives regarding the legal, social and ethical aspects for the period were the following:
▪ Completion and submission of D1.7 “Assessment of project development from social, legal and ethical viewpoint: WP3 Report”
▪ Completion and submission of D1.8 “Assessment of project development from social, legal and ethical viewpoint: WP4 Report”
Summary of the tasks carried out and significant results
Activity 1.6 Assessment of social, legal and ethical compliance during project development
Within WP1, the mappings developed in early deliverables, D1.1 to D1.4 and the methodology first laid out in D1.6 have paved the way and provided a framework for the assessment of WP3, whose results have been published in deliverable D1.7 at the end of February. This assessment has identified a number of key issues that will require further analysis in the forthcoming project phases (namely WP4 to WP6), such as, among others, the scarce evidence provided so far of security, financial, and convenience gains, and consequently, the need for better tools to assess and produce credible data, the lack of a specific legal basis for automated border checks, the problem of non-systematic checks for EU citizens, and some privacy and data protection issues.
All previous deliverables corresponding to this WP (D1.1 to 1.8) have been also considered in the design and development of the pilot test. The WP team has carried out the following tasks:
• Some findings of the ABC4EU project so far, especially regarding the design of the pilots, were presented by Eticas at the EAB Research Projects Conference (EAB-RPC) held last September, 2016 in Darmstadt and organized by European Association for Biometrics (EAB) in cooperation with Joint Research Center (JRC) and Fraunhofer IGD. The name of the presentation was: “ABC4EU pilots: evaluating ABC gates from a legal, ethical and societal standpoint”.
• ABC4EU pilots: evaluating ABC gates from a legal, ethical and societal standpoint.
• Gathering and classification of information about the ethical, technical and societal issues concerning the intervention of the different partners in the pre-pilots. A questionnaire elaborated by Eticas with this purpose and a table concerning the potential use of personal data within the pre pilots was shared with the consortia, and useful inputs were subsequently received from each partner involved.
• Gathering and analysis of information regarding the ethical implications of the ABC gates, following the specifications provided by the consortia partners. Knowledge and results from previous EU projects, such as FastPass, FRA, have also been taken into account.
• Ethical assessment of the data collection tools to be used for the pre-pilots.
• Development of the consent form for the Spanish pre-pilots, with the help and support of Indra.
• Development of a methodological design for the pre-pilots, including the structuration of the variables and the 32 indicators. These elements were incorporated into the survey and a template for non-participant observation was elaborated. Finally, the pre-pilots survey was elaborated and subsequently adapted to include questions and concerns proposed by partners.
• Testing of pre-pilots methodological tools at each location, both Lisbon and Madrid, applying a total of 25 questionnaires and conducting observations at both airports.
• Recommendations regarding ethical and legal compliance of different partners’ participation during the development of the project
WP2 TECHNOLOGY ASSESSMENT AND END USERS REQUIREMENT DEFINITION
Objectives for the period
The objectives and methodology for D.2.7 were successfully established
PwC worked in the review of the contents and the structure of deliverable 2.7 to facilitate the comprehension of the ideas expressed and favour the acceptance of the ABC system, obtaining feedback from the different end users.
After the revision of the structure of D2.7 PwC worked in the definition of a structure for supra-national coordination and made an assessment about the needs for strategic analysis and mechanisms to share information among BCPs and the National Back Office, and between the National Offices and the EU Coordination Centre.
Moreover, PwC established the methodology used to analyse the stakeholders influence and interest on the project, taking into account the workshops carried out with various partners, End-Users and government agencies.
On the other hand, as D2.7 should achieve the objective of standardization, PwC established a common set of guidelines for the coordination and harmonisation among the Governance Board. At this point in this deliverable the SNA methodology, defined by MIR-DGP, and developed as a third dimensional model by URJC, was explained as an important tool to facilitate the understanding of the network´s behaviour.
Summary of the tasks carried out and significant results
Activity 2.6 EU common basic set of functional requirements
The main result of WP2 has been the establishment of the D.2.7 requirements structure and the lessons learnt through our research and the Technical Boarding Meeting with the Consortium members. The consortium has been provided with the following results:
▪ Strategy and coordination requirements based on Governance, Security, Communication, Data Processing, and Legal, Social and Ethical concerns.
▪ Stakeholder analysis
▪ Literature review to validate the requirements
▪ Comparative results using interview skills
▪ Stakeholder analysis.
▪ Literature review to validate the requirements.
▪ Comparative results using interview skills.
▪ Strategy and coordination requirements based on Governance, Security, Communication, Data Processing, and Legal, Social and Ethical concerns.
▪ SNA methodology for a better understanding of networks.
The creation of a new board, called EEUAB (External End-User Advisory Board) was setup to facilitate the cooperation among end users that are not working for the ABC4EU project.
After the revision of the D2.7 content by EC, PwC worked in the inclusion of all corresponding changes that applied. The set of changes applied to D2.7 are described below:
▪ Clarifying the sources of the requirements.
▪ Clarifying the participants of the surveys used.
▪ Adding new literature review.
▪ Including critical assessment taking into account the referred frameworks.
▪ Modifying the deliverable with a specification of the methodology (who decided the requirements and how).
▪ Including the specific objective and tangible impact of that work. Specifying which its contribution to the project was.
Under Activity A2.6 the project partners established end-user driven functional requirements for the National Facilitation Programme and Mobile subsystems workflow through collaboration with Spanish CNP Border Guard police. The need for seamless integration in a harmonized ABC4EU border-crossing approach was a central requirement needed for designing the systems interoperability to be deployed by multiple Member States in heterogeneous BCP settings. The key goal of facilitating TCN passenger border crossing flows, was updated to follow the most recent EC proposal for the revised 2016 Smart Borders package and clarified the maximum levels of privacy and security so to align with the EU data protection regulatory framework including the GDPR that concludes all biometric data that can re-identify a person is classified as sensitive data, while enforcing the social and ethical aspects from WP1.
URJC team worked in deliverables 2.6 and 2.7 in the development of a methodology regarding the Social Network Analysis (SNA). This methodology brings forward a study to establish a model and describes the structure of the End Users community from the SNA point of view by presenting the results in a matrix of links between nodes. Different cases of simple networks for the end users have been studied and performed in the methodology as the annex shows, but in this deliverable we have chosen the cases that shall apply. Moreover URJC has considered three authority levels with many interconnections in the global network. Different virtual graphical representation of the cases studied were included.
WP3 ABC Processes & Workflow design
Summary of the tasks carried out and significant results
Activity 3.3 Biometrics: Evaluation and optimization of 2nd Gen EU eID cards and residence permits.
During the analysis phase, the following goals were achieved:
▪ Completion of assessment regarding Ergonomics and Biometrics Needs in the case of facial biometrics. All the different physical configuration was analysed, and several recommendations were made in each configuration.
▪ Completion of assessment regarding ABC for passengers with disabilities in the case of facial biometrics.
▪ Completion of assessment regarding Streamlining human interaction..
▪ Completion of assessment regarding the challenges that must be addressed when improving face image quality for biometric recognition. An experimental evaluation with real data, evaluating the influence of the illumination over the final facial image quality in an ABC environment was presented.
▪ Completion of assessment regarding the application of multimoda fusion in the case of the ABC systems.
▪ Completion of assessment regarding the state of the art in the case of the research about face image quality for biometric recognition.
WP4 EU ABC ARCHITECTURE
Summary of the tasks carried out and significant results
During the analysis phase, the following goals were achieved:
▪ A very high level (from the point of view of the traveller) of the different processes that a traveller must undergo when interacting with each subsystem and high level data entities involved in those processes for better understanding were streamlined.
▪ Requirements Catalogue: With the information generated in previous deliverables and the decisions made in the Business Process Model, the main requirements that an ABC System should address were identified, not only inside the scope of the current project, but also for their consideration in future European R&D projects
▪ Detailed model were analyzed, studying the system under different perspectives:
• Business perspective (use cases model)
• Data structures perspective (data model)
• Functional perspective (class model)
• Interactivity with travellers perspective (user interface model)
For the preparation of the pilots, the work to be done was defined in several business cases, including
▪ Graphical location of the pilot into the general process of each subsystem
▪ Definition of the scope of each pilot by narrowing the requirements identified in previous analysis, while extending the requirements catalogue with specific needs for each pilot
▪ Description of work: clarification of what outcomes are expected at the conclusion of the project, and how they would be achieved.
▪ Design models for those pilots that will implement software modules, including new point of views of the system:
• Dynamic point of view (scenarios/ State-Transition diagrams)
• Data model
• Class model – component and deployment Models
• User interface model
The design of the architecture for the different subsystems (Biometry, E-documents, RTP, EES CES, ABC interoperability, external systems interface) was performed according to the requirements. A common BMS (Biometric Matching System) for the EES and the RTP was designed, and the interaction among the different subsystems during an ABC crossing by a TCN was analysed and defined in the architecture definition. Also a C4I Integration Concept was written, which analyses how such a system could be integrated with an ABC system.
The deliverables were adapted to the requirements of the new Smart Borders proposal (released April 2016), as it affects the ESS and the RTP (which becomes NFP) mainly as the project commited with adapting the work to the everchanging legislation.
Additional research in the scope of WP4 carried out by the consortium:
▪ Analysis of current and state-of-the-art biometric solutions implemented or potentially applicable to the ABC Systems, VIS, EES and RTP. Focus on facial and multimodal biometrics.
▪ The list of functional requirements in an ABC system related to biometrics module were proposed.
▪ Analysis of the current potential weakness of the biometric module in ABC systems. Two main fields to improve the current state of the art were detected: multimodal fusion and presentation attack detection.
▪ Development of two business cases according the weakness detected: multimodal fusion and presentation attack detection.
▪ Proposal and guidelines generation of the biometric module High Class Model (logical model of the software system under construction).
▪ Several guidelines for the Signalling User Interface Model were developed considering different international standards.
Because of the general last years increment of the cyberattacks in all IT systems, the effort dedicated to antispoofing or presentation attack detection techniques has been bigger than the initially estimated in the project proposal. The cybersecurity field has suffered a great expansion during this last years at the same time the attacks are more and more complex. There are multiple examples of attacks performed during last years that was absolutely unexpected several years ago. A concrete proof of the increment of this technical necessities is the currently in process development of a new ISO standard (ISO/IEC CD 30107) focalized specifically in presentation attacks.
From the research point of view this field has not been deeply analysed in the case of ABC systems, so the project results were adapted to current social necessities.
From a technical point of view, a wider range of attacks than the initially expected and with higher technical levels were considered: printed face on a paper sheet (as baseline method); digital face displayed on a screen from digital devices such as tablets, smartphones, and laptops; 3D masks (paper, silicon, cast, rubber etc.) specifically moulded for a targeted face, inclusion of facial data in alternative poses (like t-shirts), etc..
The inclusion of these more complex attacks has consequently resulted in the necessity of the analysis, design, development and test of more complex presentation attack detection methods. URJC research team has included at the same time nor only more sophisticated algorithms, but already a multisensor (and multispectral) approach in order to study the antispoofing capability of this different visual information representation.
WP5 SUBSYSTEMS DEVELOPMENT & INTEGRATION
Objectives for the period
The scope of the pilots was drafted in order to anticipate the research and development activities within WP5.
The objectives in WP5 for this period were the following:
▪ Development and completion of deliverable D5.1 Biometric proof of concept.
▪ Development and completion of deliverable D5.3 Travel document management proof of concept.
▪ Development and completion of deliverable D5.5 RTP subsystem proof of concept.
▪ Development and completion of deliverable D5.7 EES subsystem proof of concept.
▪ Development and completion of deliverable D5.9 ABC Systems proof of concept.
▪ Development and completion of deliverable D5.11 Mobile subsystem proof of concept.
Summary of the tasks carried out and significant results
Within WP5, the Technical and Logistical aspects of the Proof of Concept in Lisbon and Spain were carried out.
Activity 5.1. Biometrics
The architecture of the biometric software provided by the partners was analyzed, in order to be integrated into the enrolment and verification processes.
The upgraded ABC System deployed in Spain started to be designed and built, with new camera system to support the research activities. A preliminary software module for retrieving and processing the video stream was developed by INDRA.
A comprehensive study of different algorithms for object and persons detection was performed by INDRA using the 3D sensor located over the eGate (mantrap) in a zenithal position. Tailgating and abandoned objects detection algorithms were developed. For detecting persons, head and shoulders are searched taking into account the difference in depth of head and shoulders and their apparent sizes. In order to detect abandoned objects, the 3D image is processed after noise elimination and background subtraction. The algorithms were successfully tested during Barajas pilot in numerous situations.
Atos proposed a common SOA architecture for RTP and EES to make use of a common BMS provided by the Dermalog AFIS. The AFIS would store fingerprints and perform biometric matching for fingerprint queries and it would store the facial images to be retrieved by ABC systems which would perform the matching locally.
For the biometric prototype Cognitec started research activities in order to improve biometric performance of face recognition.
In particular, investigations concentrated on the border control use case (camera versus passport image) with passport images of reduced quality, as they were reported. Apart from general improvements of the biometric performance in terms of False Acceptance Rate and False Rejection Rate, Cognitec was able to achieve improvements especially for images taken with imperfect lighting conditions and non-standard pose.
The results of these developments were provided to the partners in the project by means of the Cognitec product FaceVACS SDK v. 9.2.
To further improve the prototype biometric algorithms Cognitec addressed problems like compression artefacts in passport images and partial facial occlusions including glasses, which can occur with both camera and passport images. Cognitec was able to achieve some improvements here, though these are not yet consistent on all types of images present in the test bench. So, this development was considered still ongoing and required a later update.
URJC team has worked in the study and development of different Presentation Attack Detection (PAD) techniques. Different methods to detect spoofing attacks in the case of facial biometrics were analysed. After a deep analysis of the possibilities available, four types of vision sensor (RGB, IR, NIR and depth information) were selected and its application to presentation attack detection were studied.
In order to perform this research, the different antispoofing facial databases available in the research community were analysed. One of the main conclusions after this study was that currently any of these databases could apply to an ABC system particularities. Any of them consider the use of a multispectral approach, and the attacks weren’t usually adapted to an ABC environment. Considering this restrictions, a new database called Frav-Attack database was acquired, adapting the environment and the quality of the attacks to an ABC environment.
The acquisition of this new database from our Facial Recognition and Artificial Vision (FRAV) group in URJC wasn’t initially included in the proposal, but after the analysis of current state of the art attack databases, a lack of attack situations admissible for an ABC environment was detected. The final decision was to design and acquire this new FRAV-Attack database to cover this gap of quality data. It is an extra-effort but this database is highly relevant for research and experimentation in this project, and globally it is a significant contribution.
On the other hand, URJC team researched in the development of optimal Multibiometric fusion techniques for ABC gates. The methodology to analyse specific parameters in ABC systems (nationality, age, date of issue of documentation, etc) and how they can affect biometric subsystems were developed.
For the biometric prototype Dermalog focused on increasing the customisation possibilities and stabilisation of interfaces. Additional effort was brought to the automation of the Web AFIS installation.
The results of these developments were provided to the partners in the project.
Dermalog’s “Web AFIS” biometric prototype represents an AFIS (Automatic Fingerprint Identification System) that provides a centralized Web service interface.
The pilots developed by INDRA, ATOS and VISION-BOX for ABC gates and mobile verification solutions connect to the database going through this “Web AFIS“, which then enables access to the required operations on biometric and/or demographic data.
During this period, Dermalog also tackled research activities in order to improve the reliability of fake fingerprint recognition:
▪ significant increase in sample size of real humans’ and fake fingers (including increase in variability of fake materials)
▪ enhancement of fake detection support for optical as well as capacitive fingerprint capturing technology
▪ extension of fake detection to support different fingerprint scanners (up to now ZF1, LF1, ZF10, LF10)
▪ substantial improvements of fake finger detection classification algorithms (significant reduction of FRR and FAR)
▪ integration and presentation of fake detection functionality by means of demo applications
▪ development of quality testing and evaluation procedures for fake detection functionality
The results have been provided for the ABC4EU project as a plug-in for the LF10 scanner.
Parallel research activities focused on improving verification processes in the context of aging, to increase the accuracy of fingerprint recognition for persons under 12 years of age. Studies were carried on fingerprint growth and the persistence of fingerprint recognition when aging.
During the course of the project, Dermalog identified an issue in which users at border controls are often confused with the use of fingerprint scanners and passport scanners, and mistakenly place their fingerprints on the passport scanner and their passport on the fingerprint scanner.
It was decided to tackle it by developing a new multipurpose scanner designed for national registration and automated border control especially suited for integration into ABC Gates, able to capture both fingerprints and passport images on the same scanning surface, with the aim of facilitating end-user experience and increasing efficiency and speed of document and identity checking.
Concrete actions during this period have been:
▪ Development of a prototype for a multi-scanner
• Alpha-Release C und .Net SDK for internal Dermalog integration of the VF1 multi-scanner
• Beta-Release C und .Net SDK for external integration of the VF1 multi-scanner
▪ FBI certification of the VF1 multi-scanner
The multi-scanner offers the following features and characteristics:
▪ Captures fingerprints with highest quality and liveness detection/fake finger detection
▪ Scans and reads passports with full color scan, IR, UV and RFID
▪ Analyzes optical security features of the passport
▪ Capturing of signatures (with additional accessory)
▪ Scans passports, ID cards, photographs, flight tickets, arrival/departure cards and other small documents
▪ Reads RFID-Smartcards
▪ Automated detection of fake passports (optional)
Activity 5.2 Travel Documents Management
A vendor-independent security patterns database and the corresponding engine for validating travel documents was designed, and a preliminary version of the engine was implemented.
Later the CES was implemented according to BSI specifications (TR_03129-2).
We would like to highlight that during the study of the BSI specification, INDRA found what seemed to be an inconsistency in the specification. ABC inspection systems should be able to obtain a DS certificate from an external source if a particular eMRTD EF.SOD does not contain the DS certificate, using the SignerIdentifier found on the eMRTD. BSI specification refers to this possibility and it even defines alternate behaviour in one of its WS operations, GetDocumentSignerInformation(), in order to cover this case, but it incomprehensibly misses the necessary output parameter for returning the DS certificate (BSI specification “typo”?).
In fact, this functionality is recommended by FRONTEX best practices and it was proposed to add it to the ABC4EU CES, because, contrary to what was initially supposed, it is not uncommon to find modern passports without a DS in the chip.
Indra informed the BSI about the inconsistency in their specification and they acknowledged the issue. A conversation with BSI followed in order to discuss possible workarounds.
The CES could be considered to be improved in this way for the Final Pilot to create an online DB of DSCs as it is clear than many passports do not include DSC in the SoD. From piloting mobile metrics Atos observed 53% of ePassports did not contain the DS certificate, so providing access to DS Certificates online would make a big impact on validating eMRTD data.
The eMRTD validation subsystem (TDMS) was completed by INDRA, and used for Barajas PoC in: 1) Indra verification kiosk, 2) Indra EES enrolment kiosk and 3) Atos NFP/EES Enrolment Station.
Regarding the Mobile System, Atos worked closely with the French mobile device supplier to improve the interoperability, stability and functionality of the supplier’s e-MRTD reading and authentication SDK which is used by Atos ABC4EU Mobile prototype app, (including a feature roadmap for implementing Passive, EAC and Active authentication for piloting). Much effort was devoted to perform interoperability testing and troubleshooting by Atos for ePassports from around the world for Passive and Chip authentication. Additionally, the mobile has been implemented to make use of either a local CSCA DB or make use of CES for online validation of the CSCA in Passive Authentication.
Activity 5.3 Registered Traveller Programme subsystem development
Atos started the software development on the prototype RTP Enrolment Station (ES) and back-end RTP service. However, in April 2016 the EC released a revised Smarts Border Package which dropped the RTP central service and replaced it with a voluntary National Facilitation Programme (NFP) for facilitating registered travellers. Atos spent a month analysing and agreeing with partners the impact of the revised package, and all partners agreed in modifying the scope of the RTP delivery to follow the new NFP specification for registered travellers. It was also agreed that Atos would additionally expand the scope of the Enrolment Station so to provide a common EES & NFP Enrolment Station instead of just NFP. An updated EES registration interface was re-implemented.
The NFP development results consisted of a combined EES and NFP Enrolment Station (ES) web application, NFP backend service, NIST Pack backend service, Front End Gateway (FE GW), Kibana Logstash audit & event log system. The ES registers VE & VH TCNs and their biographic and biometric data in EES before proceeding to optionally register the TCN (biographic data) in the NFP registration.
The Enrolment Station integration was carried out with EES and NFP systems for enrolment in these services and also with CNP gateway system as it was needed to authorise the BG access (authentication in pilots under real operational scenarios) and also do background checks before enrolling in NFP. Later technical DB issues in Argos resulted that during PoC piloting the background checks were only carried out for ABC verification and not ES enrolment. The ES integrated the Indra eMRTD reader application for reading and verifying physical security elements of ePassports and Visas.
Activity 5.4. Entry/Exit subsystem development.
The EES was reviewed and redesigned (according to the new Smart Borders proposal).
Activity 5.5. Existing ABC upgrades (Airport, land, sea).
The new legislation proposal was also taken into consideration for the roadmap of the ABC setup to be tested in both the PoC and Final Pilot. The roadmap included the main features to be included in both the Enrolment Station and ABC eGates.
The scope for the different pilots was defined and the usage workflows were specified for the different scenarios and ABC topologies to be covered by pilots (mantrap or two-step segregated, TCNVH or TCNVE, first time crossing or not, registered in the NFP or not, etc.). For each scenario it was analysed how the NFP and the EES would have to be integrated. The software modules in existing Indra ABC system that would require modifications to communicate with NFP and EES were identified. It was also analysed how the current legislation and the current implementation of existing systems affects and constraints the implementation of the pilots. The specific hardware devices for pilot implementation were chosen (tablets, document readers, fingerprint readers, cameras, PCs, etc.).
Atos helped define the validation process of TCNs according to the revised Smart Borders package, with particular focus on the design of TCN workflow for visa exempt and visa holder TCNs so that they could be enrolled (using ES and kiosk) and proceed to be validated at the border by using upgraded ABC solution (or mobile information units). Atos also developed the Front End GW to enable easier integration with different vendor ABC WS and Restful interfaces, and thus supports interface harmonization towards back end services such as NFP enrolment status checks.
A complete ABC software suite was developed (verification kiosk, enrolment kiosk, mantrap and ABC monitor) integrated with VISIONBOX EES and ATOS NFP system. The biometric statistical information to be logged by the system was agreed with URJC and UMIL. The effort was first aimed at the mantrap system to be used in Barajas. Later the extensions needed in the ABC software in order to accommodate Algeciras pilot features (segregated two steps with facial recognition as the token for automatic door opening) started.
The inclusion of a presentation attack detection evaluation was performed during the pilots test. Because of the last years increment of the problems with cybersecurity attacks, the number of attacks was bigger than the initial expected in the proposal.
Activity 5.6 Mobile subsystem development
The mobile development activities consisted of developing an Entry Exit Service verification workflow for reading ePassport data, reading Visa data (including OCR capture with camera, as well as swipe reader, especially suitable for visas not properly affixed to passport), taking fingerprints for matching records in EES (VE) and VIS (VH) checking existing EES records, checking backend national and central DBs including VIS and SIS (via CNP Argos system) before electronically stamping the border crossing in EES or referring to 2nd line (manual booth). Authentication access control is implemented though integration with CNP Argos system. The mobile subsystem additionally checks the TCN´s NFP status so to allow BGs to achieve significant time-savings for NFP-enrolled TCNs (as the NFP pre-vetting checks avoid the need for border guards to carry out SBC thorough checks).
The Mobile subsystem proof of concept implementation was made available. Updated EES functions and interfaces were re-implemented.
Activity 5.7. RTP, EES and upgraded ABCs integration
The following tasks were carried out:
▪ Integration with the initial EES.
▪ Integration of Dermalog decadactilar scanner in Indra EES enrolment kiosk.
▪ Integration of Dermalog unidactilar scanner in Indra verification kiosk.
▪ Integration with Cognitec’s 1:1 facial recognition software.
▪ Integration with Atos NFP RESTful service.
▪ Integration of the CES (BSI interface) with Atos FrontEnd server.
▪ Integration of the TDMS in Atos NFP/EES Enrolment Station.
▪ Integration from the Atos Enrolment station and Mobile System over the Atos FE GW towards the initial EES.
▪ Integration from the Atos Enrolment station and Mobile System via EES with Dermalog BMS via the EES with correct image format.
▪ Integration of Dermalog decadactilar scanner in Atos NFP/EES Enrolment Station.
▪ Integration with Cognitec’s facial capture software in Atos NFP/EES Enrolment Station.
▪ Integration of the Indra iBCS eMRTD desk top application in Atos NFP/EES Enrolment Station.
▪ Integration from the Mobile System over Atos FE GW to the CES (BSI interface).
▪ Integration from the Atos Enrolment station and Mobile System over the Atos FE GW towards with CNP Argos Authentication web service.
▪ Integration from the Atos Enrolment station and Mobile System over the Atos FE GW towards with CNP Argos pasoFrontera web service, and checks to national DBs, VIS and SIS with correct NIST Pack formatting for VIS checks.
▪ Integration from the Atos Enrolment station and Mobile System over the Atos FE GW with the Atos NFP service.
▪ Integration of Mobile System with Coppernic API, IntgeratedBiometrics Fingerprint API and Zebra OCR API.
▪ Integration from the Atos Enrolment station and Mobile System towards kibana Logstash event logging system
WP6 BCP Test & Validation
Objectives for the period
With the advent of the PoC execution, the main objective of the period was the definition of the validation framework.
Summary of the tasks carried out and significant results
The consortium has been provided with the following results:
▪ Validation framework divided into: design, performance, and analysis phases
▪ Limitations to take into account from a technical, ethical and operational point of view.
▪ Literature review to validate the framework.
▪ Comparative results using interview skills.
The validation framework was written in its final draft version where:
▪ A Validation Framework was proposed in order to serve as guidance for the team responsible for carrying out the pilot activities during the ABC system development.
▪ The goals and guidelines with regard to the design, performance and analysis of the phase of the ABC system, regarding processes and protocols, were defined.
▪ The responsibilities for defining, executing and analyzing the results of all validation activities to be performed throughout the project were established.
On the other hand, the consortium worked in the first draft of deliverables 6.2 6.3 and 6.4 which are focused on ABC, EES and RTP concept and validation plans.
The validation activities for each system are related to:
▪ Ergonomics.
▪ Document verification capabilities.
▪ Biometric Identification.
▪ Access control and perimeter surveillance.
▪ Integration with external systems (VIS, API, SIS, EES, RTP, etc.).
▪ Mobile/integrated solutions.
▪ Central control and support systems.
Opinions, experiences and feedback provided in the last EUIWG and EEUAB meetings (9-10th of May) were incorporated in terms of: ergonomic, functional and operational requirements, use cases, EES new regulation from the European Commission, etc.
PwC prepared the structure of validation results deliverables aimed to gather the results, information and conclusions obtained during pilot phase.
Likewise, concerning the PoC, PwC elaborated a set of forms to facilitate the collection of data during the pilot phase. Furthermore, PwC gave support to ETICAS with activities related with travellers’ feedback. These forms were focused on gathering the following main information:
▪ ABC: ePassport placement time, ePassport reading time, time for facial image capture, time for fingerprint capture, time for verification process and attempts.
▪ EES: ePassport placement time, ePassport reading time, time for facial image capture, time for fingerprint capture, time for BG authentication process and attempts.
▪ NFP: ePassport placement time, ePassport reading time, time for facial image capture, time for fingerprint capture, time for verification process and attempts.
▪ Mobile device: ePassport placement time, ePassport reading time, time for facial image capture, time for fingerprint capture, time for verification process and attempts.
▪ Others: rejects, failures during processes, feedback, age range...
Moreover, two EEUAB meetings has been prepared by PwC for the attending (online connection) of several end-users for sharing the approach, results, tests and experience obtained during the PoC.
Atos provided contributions to Activity 6.3 with revision of earlier validation plans to take into account the changes needed for NFP as it superseded RTP, adding mobile system details needed for the mobile validation proposed metrics to be obtained during piloting for NFP Service, Mobile System and Enrolment Station. As one of the main integrator/supplier of piloted systems Atos have participated in and supported on site on a daily basis the validation testing with real end-users, travel documents and TCNs of NFP Service, EES & NFP Enrolment Station and the Mobile System at International Madrid Airport piloting site.
Mainly, the Proof of concept execution activities consisted in inviting TCNs in possession of electronic travel document (VE and VH) with support of pilot partners and AENA staff to sign consent form and enter pilot flows. First step is to be enrolled on EES and, optionally, NFP at the Enrolment station, located near the border crossing booths but in a specifically signalled area sufficiently remote to avoid interference with the non-pilot flow and/or queue, and subsequently border guards invite the TCNs to test ABC4EU border crossing using eGate or Mobile System. During mentioned EES and NFP enrolment, the Atos system was recording the processing times of all the relevant events taking place in the Enrolment Station during the process. The selection of these events was informed previously to PwC (in charge of the validation process) by Atos, prioritizing the complete process of enrolment in EES as the most important metric to be analysed since it is in this step that EES thorough checks are conducted, which would save time for the TCNs in subsequent border crossings.
Mentioned enrolment was carried out by a border guard (BG) from CNP, who follows the screen instructions to validate the travel documents, capture fingerprints (optionally capturing face with reflex camera when passport image lacks sufficient quality) and perform thorough checks. The Enrolment Station of Atos is designed to process both short stay visa holder and visa exempt TCNs, meaning a majority of TCNs could be enrolled and, after successfully finishing this enrolment phase, each TCN was accompanied by the BG to either eGate or Mobile system. There, the border crossing happens and EES is updated from the eGate or mobile, having first validated the travel documents and checked status of EES and NFP records for this TCN.
The first proof of concept validation stage aimed at performing an end-to-end testing of all the involved research activities, fully integrated in a real deployment scenario. The main objective of the validation testing was to assess the results of the third country nationals processing through an automated border control, as it could be implemented in the future, including additional functionalities to the currently deployed systems which were modified for the purpose of executing the activities planned in the project.
The results of the research activities were fully integrated and tested (CES with BSI interface implementation, new TDMS travel document physical security features validation, tailgating detection, lost object detection, new biometric algorithms, EES and NFP). By testing with real travellers in a real border control scenario, where the Smart Borders regulation was not approved, the systems were adapted to accommodate to the production police systems and to comply with the currently legislation in force.
The expected outcome of this preliminary proof of concept was:
• Assessment of the interoperability of all the elements deployed,
• Assessment of the impact on the current processes of third country nationals enrolment at the borders,
• Assessment of the impact on the current processes of third country nationals automated processing,
• Assessment of the suitability of a NFP functionality to fully automate thorough checks of third country nationals,
• Assessment of a mobile device aimed at helping the border guard for manual processing,
• Assessment of a new travel document physical security features validation engine,
• Assessment of new algorithms for tailgating detection, both in one step and segregated two steps topologies,
• Assessment of new algorithms for lost and abandoned objects detection,
• Assessment of the use of facial verification as a token for segregated two steps topology,
• Assessment of interconnection and interoperability between different enrolment and verification schemes (self enrolment+eGate, self enrolment+mobile, enrolment station+eGate, enrolment station+mobile),
• Assessment of ergonomics and human-machine interaction (self enrolment, travel document handling, self biometric capture),
• Assessment of the impact on the current border guard processes and systems.
URJC developed an evaluation framework oriented to the measurement of the system response to antispoofing attacks. This theoretical approach was adapted to ABC particularities, considering the pilot architecture.
The main tasks developed during the pilot were:
▪ Several antispoofing attacks were emulated in the real pilot, studying the influence of several parameters over the presentation attack detection capability of the system. The attack considered and performed in the pilot were:
• Printed face on a paper sheet. Sometimes a printed face is shown with eyes cropped out so that the subject’s eyes blink underneath.
• Digital face displayed on a screen from digital devices such as tablets, smartphones, and laptops. This kind of face presentation can be static or a video. In video attacks facial movements, eye blinking, mouth/lip movements or expressions are simulated.
• A paper 3D mask, emulating the response to the system to a close to real texture with a different 3D shape.
• A high quality 3D mask (silicon) specifically moulded for a targeted face.
• Images attack camouflaged in some part of the clothes, like a shirt.
▪ A systematic observation of the travelers and it interaction with the registration and verification kiosk were performed. The standard model “human-biometric-sensor interaction (HBSI) evaluation method” was adapted to the ABC scenario defining some metrics.
The number and technical level of the presentation attacks considered has been bigger that the initially estimation, and the effort dedicated to this task is slightly bigger than the estimation in the project proposal. This increment of pm doesn’t imply an increment of the global EC funds to URJC because the current URJC pm cost is lower to the initially supposed. The final team isn’t as senior as it was initially expected in the proposal, so the pm cost is lower.
Third period (2017-2018)
WP1 LEGAL, SOCIAL AND ETHICAL ASPECTS
Summary of the tasks carried out and significant results
In the framework for Deliverable 1.9 the following tasks were conducted:
• Updating ABC legal frameworks: data protection and automated border control.
• Adapting ethical assessment methodology to WP5 specifications and requirements. Eticas studied the integration and deployment of the Registered Traveller Programme and Entry/Exit System within the ABC4EU project and the concept for upgraded ABCs according to each type of border and passenger requirements.
• Subsystems analysis: data life cycle reconstruction, examination of performance and identification of vulnerabilities. This assessment focused on the relative data protection compliance of these systems, privacy concerns derived from sociotechnical interaction and their data management protocols when deployed. In this context, particular attention was given to the social impact of biometric processes.
• Constructing recommendations on this basis
In parallel with these assessments, three main groups of tasks were carried out within the package:
- Tasks concerning logistical aspects of the project (including organizing and attending internal and advisory board meetings and the elaboration of reports about the project´s evolution);
- Tasks supporting project partners in tackling ethical aspects of the field research (including the development of consent forms and analysis of the materials provided by other partners with a data protection focus);
- Tasks for dissemination (including publications and attendance to conference and workshops).
Deliverable 1.10 includes assessment of the pilots’ execution as part of the pilot evaluation carried out in WP6, in order to complete the examination of the ABC system implementation and to measure its potential social impact. In this context, Eticas designed and applied a survey and a number of non-participant observations in the various locations (sea border, land border and air border locations) where the system was pre-tested and tested. The methodological approach developed by Eticas for the fieldwork focused on the experiences of the travellers, their opinions and expectations. It was constructed considering a wide set of variables about acceptability, desirability, functionality and adaptability to vulnerable groups. Moreover, the survey was designed to consider the different inputs offered by the project partners and integrated information on variables required by them within the data collection tools.
217 surveys were administered during the final test phase at the Lisbon airport, Algeciras port and Calafat land border. The results of this final research process, as well as the 39 surveys and non-participant observations carried out during the pre-test phase, are reflected in D1.10. Besides this, the document includes an analysis of these findings contrasting them with the information about the system validation provided by the project partner PwC and the results of the entire pilot tests, including the ones conducted in Algeciras. The analysis of all this relevant information was also analysed against previous projects conducting field research in this area (FASTPASS, BODEGA, EU Lisa and others), as well as considering an analysis of the border guards’ opinions on the current transition towards an automated border system. The Deliverable thus constructs preliminary and explanatory correlations about the acceptability of the proposed solution against a selection of sociodemographic variables.
Regarding D1.11 the final report on the social, legal and ethical implications of the ABC4EU project includes a thorough review of our findings throughout the project. It also includes a systematic review of the legal state-of-the-art as it affects the ABC4EU system. This was particularly important considering that the legal framework underwent a significant number of important changes during the project' timeline. An analysis of the labour rights of the border guards and how these might be affected by the implementation of the ABC4EU system was also carried out during this period. Finally, a conceptual framework for interpreting the full range of the social, legal and ethical implications of the ABC4EU system was developed and incorporated into the report.
To conclude, Eticas carried out many tasks concerning the ethics compliance of the project fieldwork. The work of Eticas in ABC4EU has been mainly driven towards building the legal and ethical references to be taken into consideration for its development and towards monitoring the compliance of these principles and regulations in terms of the technological design. The tasks developed in this context included:
a) the design, translation and adaptation of consent forms for all the pre-pilot and pilot tests,
b) the co-design, together with the project partners, of the notification about data management specifications for each pilot test to the Data Protection Authorities in Spain and the monitoring of these notifications for all the countries where the tests were conducted (Spain, Portugal and Romania), and
c) the design and gathering of a non-disclosure agreement to be signed by all the partners involved in the Calafat pilots, which was expressly requested by the Romanian Police during the conducting of fieldwork in this country.
These protocols aimed to guarantee that all partners’ interventions in the development of the pilots was in line with data protection and human rights regulations and principles. The Annex of D1.10 includes a thorough explanation and refers to other contributions made in terms of protocols for human-machine interaction in the system´s implementation, the compliance of fieldwork activities with informed consent protocols, the minimization and deletion of personal data and putting safeguards in place for the treatment of all actors involved.
WP2 TECHNOLOGY ASSESSMENT AND END USERS REQUIREMENT DEFINITION
Summary of the tasks carried out and significant results
The D2.8 is intended to be an EU Border Guard common set of requirements for Strategic and supra-national coordination. Specifically, it aims to provide information about needs for strategic analysis and mechanisms to share such information between the BCP and the national back-office, and between national offices and an EU coordination centre.
These goals were already mentioned for D2.7 but in this case it was the second iteration. Concerning D2.8 the purpose pursued was to enhance and enrich the set of EU border common guard requirements for strategic and supra-national coordination presented in deliverable D2.7 with EU-wide application including all types of BCP, e.g. airports, land and ports.
It is important to highlight that the set requirements were collected by considering regulations, frameworks, international standards and projects as a reference to support them at each level of the final Governance Board structure defined in D2.8.
As second major input, since WP6 activities had to be performed in parallel, PwC decided to take advantage of the information generated during pilots for this deliverable (pilot experience). In this regard, it was possible to obtain information from different sources such as feedback provided by the border guards (CFRONT, CNP and SEF), checklists developed for the first PoC, logs collected by the technical partners from different systems' topologies (the systems' deployment depended on the pilots location), and observation by PwC team.
In addition, regarding WP6 activities, it was gathered all the information obtained from the meetings held with stakeholders and end users, considering the BG approach. In this regard, the meetings conducted were focused on planning the PoC activities.
The deliverable 2.8 meant two major challenges. On the one hand, it was necessary to investigate and obtain accurate and wide information referred to strategic and supra-national coordination. On the other hand, PwC had to perform the WP6 activities in parallel to this deliverable (so it was decided to take advantage of the pilot experience for the D2.8 as mentioned before).
WP5 SUBSYSTEMS DEVELOPMENT & INTEGRATION
Summary of the tasks carried out and significant results
The goals for this period were to achieve the development of final solutions to be deployed in the final Pilots, comprehending:
o Final Biometric Management System
o Final Travel Document Management System
o Final Entry-Exit System
o Final NFP System
o Final ABC System
o Final Mobile System
Within WP5 (Subsystems Development & Integration), research and development activities have been carried out in order to improve the biometric algorithms taking into account the results from the proofs of concept performed at Madrid, Lisbon and Algeciras. These algorithms, which include biometric fusion, facial anti-spoofing, tailgating detection, lost objects detection, and facial recognition on the move for two-steps topology, were fully integrated into the final prototypes. The CES is being extended with an innovative feature that pursues that an e-passport can be trusted even if the necessary master list is not available. The EES has been extended by adding AFIS capabilities and other features. The ABC prototypes have been extended to cover additional use cases (like reading electronic residence permits). The mobile device, the NFP and the Enrolment Station were also reviewed and extended for the final pilots.
Activity 5.1 Biometrics
After Algeciras PoC, Indra agreed with UMIL, URJC, Cognitec, Dermalog and Atos the functionality for the biometric fusion algorithm. UMIL developed two versions of the algorithm in collaboration with Cognitec, Dermalog and URJC. A basic version only takes into account the facial and fingerprint verification scores, and a second, more advanced version, takes as input also the quality of the images. The algorithm for calculating the quality of facial images was designed by Cognitec, while the method for getting the quality for fingerprint images was provided by Dermalog.
UMIL has focused its research in the analysis and development of biometric multimodal fusion methods suitable for ABC scenarios, including the investigation of different fusion strategies, performance evaluation procedures, and early tests ran on biometric data representative of ABC conditions. As a first step, UMIL has studied the suitability of fusion methods at different levels, concluding that score level fusion methods and hybrids of score and feature level methods are the most suitable. Among these, UMIL implemented and tested seven different methods. The methods were chosen paying attention not only to their effectiveness, but also to the possibility of using them in a privacy compliant way. Among the tested methods, UMIL considers as good solutions the application of two final fusion methods detailed in the deliverables. UMIL has also provided illustrative implementations in C language of the selected methods preliminary trained on the biometric match scores computed on an external public dataset using the same biometric matching software used in ABC systems. Preliminary normalization of the scores were also considered. In addition, UMIL has provided suggestions on which thresholds can obtain a good performance, based on the training results. UMIL has also proposed a procedure for increasing the biometric recognition accuracy in ABC systems with hierarchical biometric recognition.
URJC research team has been mainly focused in the improvement of current ABC systems for presentation attack detection in facial biometrics. Different methods to detect spoofing attacks in the case of facial biometrics have been analyzed and an initial proof of concept has been developed that can be modified to be included in the ABC gate. After a deep analysis of the possibilities available, four types of vision sensor (RGB, IR, NIR and depth information) were selected and its application to presentation attack detection were studied. Considering the restrictions and specifications of an ABC gate, it was decided to work with still images.
Two different approaches were researched for the purposes of presentation attack detection, texture-based and geometrical methods. A wide range of attacks with different technical level were considered: printed face on a paper sheet (as baseline method); digital face displayed on a screen from digital devices such as tablets, smartphones, and laptops; 3D masks (paper, silicon, cast, rubber etc) specifically molded for a targeted face.
Apart from participating in the discussion of fusion algorithms, Dermalog carried out the following activities:
• Research and improvements of the fake detection technology.
• Research and improvements of “Aging” technology – fingerprint matching under 12 years.
• Development of a new release of Web AFIS with improvements on the performance and the interface to other systems (e.g. BMS), and including technologies such as the new fake detection technology and the “Aging – Fingerprint matching under 12 years” and.Testing of the new VF1 multi-scanner
The tests done during Barajas pilot allow a valuable experience in order to improve the PAD algorithm and a more advanced approach was carried out by including presentation attack detection on the fly. This new algorithm was integrated in Algeciras final pilot.
The user interaction with the mobile system was deeply analysed during the Romanian pilot, measuring the border guard experience using and adapted methodology based on biometric user interaction.
Indra worked on improving its own algorithms (lost object detection, tailgating detection and facial recognition on the move for the two-steps segregated topology) taking into account the feedback from the PoC. Calibration, configuration and logging tools have been extended so that the algorithms can be fine-tuned easily.
Activity 5.2 Travel Documents Management
The operations needed for Terminal Authentication (GetCertificateChain() and GetTaSignature()) were implemented in the CES in order to prepare it for reading fingerprints from travel documents. The CES server at El Escorial was updated and tests were done with Spanish preproduction passports lent by CNP.
One of the main outcomes of the PoC in Algeciras was the possibility to extend the CES with a functionality that would enable to trust a passport when the CSCA certificate is missing (as is the case for Morocco and some other countries). The ground for this proposal, which was defined in collaboration with Atos (interested in integrating it both in the mobile solution and in the Enrolment Station) was the vast majority of travelers observed during the PoC pilot testing (Moroccan) coming from a country of origin which did not provide any kind of certificate required to validate the authenticity of the electronic and cryptographic material of the passport. The problem is, when this situation happens at a border crossing point, the possibilities of automation are minimal (almost 100% of the travelers in Algeciras are from Morocco). A new operation in the CES (PossiblyValidDS()) has been implemented by Indra. This innovative feature has been evaluated in the final pilots, resulting in a proposal for harmonization that will help automation under this kind of situation. Indra took advantage of this feature, which has been integrated into the Verification Kiosk and Atos did the same in the Enrolment Station (for validating a DS certificate as a candidate to be trusted) and in the Mobile System (for passport verification).
A management console for the security patterns database was developed by Indra, which allows editing the patterns with a user-friendly GUI.
Indra also replaced the current electronic documents inspection system with the CES development in the Indra kiosks. For the final pilots testing and validation, Indra kiosks and Atos Enrolment Station (and the Mobile device, as in the PoC) used the BSI interface provided by the CES, including the PossiblyValidDS() extension.
Activity 5.3 Registered Traveller Programme subsystem development
Atos has continued activity 5.3 focusing primarily on the PoC pilot results and feedback from the BG to improve the Enrolment Station (ES), the user experience and time needed to enroll TCNs in EES and NFP. Therefore, the ES GUI was completely redesigned to have a more intuitive look and feel as well as separating logical functions eMRTD & Visa Reading, EES and NFP into separate tabs instead of the one common application form.
Additionally there were new functions added as outlined below:
• authorization token for access to all backend services controlled by the Front End GW
• added English language capability
• web server functionality improvement
• added details query for TCNs previously enrolled in EES and show their enrolled details.
• added duplicates check in EES and ability to store new eMRTDs in the case that a TCN has an additional eMRTD that can be registered
• added EES crossing history query
• implement new CES service to validate DSC certificate to be saved in a trusted CES DB store for ePassports issued by countries that do not make public their CSCA certificates.
• added NFP details query to show TCNs data previously enrolled
The FE GW was updated to provide a local authorization check to all back end national and central services, so that only BG police previously authenticated against national police systems will be authorized to access a service behind the FE GW. The FE GW was also updated for all new EES, CES & NFP services and updates to existing ones as they were made available by each partner.
The NFP Service was updated for more management functions to be able to read the TCNs file and check if the status is Rejected or Approved as both are supported in the Final Pilot NFP lifecycle.
• EES & NFP Enrolment Station (ES) with its complete re-design to have a more intuitive look and feel and separation of logical functions eMRTD & Visa Reading, EES and NFP into separate tabs instead of the one common application form. All new functionality and service queries have finished implementation as described previously in the previous interim report.
• FE GW updated to implement all new and updated services towards the EES, CES & NFP subsystems, and an authorization token for controlling access to all national and central EU services over the FE GW.
• NFP Service updated for new service to read the TCNs file and NFP lifecycle management for rejected applicants.
Activity 5.4. Entry/Exit subsystem development.
Activity goals focused on addressing the PoC pilot execution results and including also additional features that were not included in the first version.
The main findings from the Proof of Concept trial execution was that fingers collection process needs to accommodate cases where the slap capture is not possible, namely
• People with amputated fingers or with medical conditions that require that the system allows them to be enrolled even without providing fingerprints.
• People with temporarily medical conditions or immediate difficulties that do not allow them to enrol their fingerprints need to be allowed to register even without providing their fingerprints with the condition that they will have to provide them on the next Entry.
The final solution also had additional features that were not implemented and tested in the POC Pilots, namely:
• Duplicate search on existing accounts:
o The final solution that was developed, deployed and piloted in the final trials is person centric, meaning that each Passenger has unique account to which multiple Passports and Visas can be associated. This solution evolved from the document centric approach, in which the same Passenger with multiple documents could create different EES accounts, each associated to each of the documents.
• Enrolment Process Supervision
o A monitoring station was developed in order to allow the enrolment process supervision by the Border Guards. This monitoring station aimed not only to follow the Passenger’s enrolment process results but also to allow the Border Guard to intervene in the enrolment process and support the Passenger’s in need.
▪ For the cases where the Passenger was not able to provide fingerprints, the Monitoring station allowed the supervising officer to exempt them to provide fingerprints. This exemption could temporary or permanent, depending on the Passengers case.
• Entry and Exit Crossing:
o The pilot of EES final solution was deployed both Arrivals and Departures, meaning that both Entry and Exit records were registered, which brought the opportunity to develop, deploy and test new features, namely:
▪ Verification of Entries without Exit - In the event of a new Entry of a pre-registered Passenger, EES assesses if there were any previous entry occurrences that do not have an Exit Record associated to it.
▪ Passenger Border Crossing History – EES system was also able to present more information to the Border Guard, namely the history of previous transactions, to help assisting on the Border Control Process decision.
Activity 5.5. Existing ABC upgrades (Airport, land, sea).
Activity goals focused on addressing the PoC pilot execution results and preparing the ABC systems to be in compliance with the adaptations required for EES Final Prototype.
The use case considered for this Final Solution was based on the harmonized automated border control processing of Third Country Nationals Visa Exempt and Third Country National Visa Holders.
The objectives for this activity were the following
▪ Ensure that the Updated ABC System Final Prototype developed during the project complied with all technical and operational requirements.
▪ Test the system performance.
▪ Provide an overview of how the updated ABC System would operate in a pre-operational environment.
▪ Identify possible limitations, risks and exceptions in the mobile inspector.
▪ Provide testing results to set of recommendations to be used as the basis for an E.U. harmonised/standard design and usage of mobile systems.
Indra worked on integrating new hardware prototypes:
• New automated doors. After asking different gate vendors for changes in their communication protocols in order to accommodate to Indra’s requirements, different doors were integrated and tested in order to harmonize the interfaces for the integration of the gates with the software.
• New facial on-the-fly verification module. This piece of hardware, which is in charge of facial recognition, has been designed in a way that will allow to be used in different configurations. A preliminary prototype was built and tested integrated in a gate. Indra worked on a second prototype, with the form factor of a kiosk, which was tested in the final pilots in a segregated two steps topology. The software for controlling the new facial verification module was also developed.
Indra was working on several aspects that were not covered in the PoC:
• Integration of tailgating detection algorithms. Although preliminary versions of this algorithm were tested at the PoC, they were not integrated with the ABC software during the first pilots.
• Integration of biometric fusion algorithms developed by UMIL. The fingerprint will be asked only if the facial recognition was both not good enough and not too bad.
• Integration of anti-spoofing (presentation attack) algorithm developed by URJC. This presentation attack algorithm has been tested in Algeciras pilots, taking the Calafat pilots as the baseline for the comparison of the results.
• Reading visa stickers at the EES enrolment kiosk.
• Integrating new EES functionality, like performing an AFIS search before enrolling a new document for avoiding duplicates (passport renewal, multiple nationalities, etc).
• The integration of the new Dermalog VF1 two-in-one (eMRTD + fingerprint) scanner has been evaluated, but not accomplished. The intention was to demonstrate the integration of a state of the art equipment, which could greatly facilitate the enrolment of third country nationals, while reducing the size of the systems at the same time. However, the device was not mature enough and its integration was withdrawn during the period.
Activity 5.6 Mobile subsystem development
Atos commenced activity 5.6 by reviewing thoroughly the BG feedback and resultantly have re-designed the GUI to much improve the BG experience and make items of attention (e.g. alerts) more obvious while reducing clicks and improving processing times by querying external DBs in parallel, and thus sending queries as soon as the required data is available rather than having a more process flow approach as implemented in the first PoC. To facilitate this, the GUI design will now make more use of tabs to enable this user experience, similar to what was observed in the Mobile Pass application. In the Final Pilot application, the BG decisions are now only needed upon initial reading of the eMRTD (so to continue the process or to end it and refer the traveller to the manual booth or enrolment or refer to 2nd line), and at the final step to allow or refuse the traveller to cross the border or refer to second line. Further to BG and traveller user experience, improvements have been made also in presenting a graphical view of fingers stored in EES and VIS available for checking against.
Additionally there are new functions added as outlined below:
• Offline mode and synchronisation of Offline Traveller Records
• New EES Exit workflow
• Biometric Fusion
• eMRTD reading extended to eRP and eID cards (ICAO compliant)
• New CES responses for online Passive AuthN indicating CSCA not available but the DSC is found in Trusted DSC store
• Face matching service (Cognitect FaceVACS)
• New Authorisation Token for access to all backend services & DBs
• New 3rd party SDKs tested:
o New Imense Camera OCR reading of Visas and Number Plates
• Updated FE GW Interfaces:
o Authorisation token to back end service & DBs
o Certificate Exchange System for Online Passive Authentication (BSI TR-03129-2 )
o Entry Exit Service (EES) updated and new services incl. EES Crossing History, 1:1 FaceMatch
o NFP Service updated for new services
Activity 5.7. RTP, EES and upgraded ABCs integration.
Activity goals focused on addressing the PoC pilot and final pilot execution results and preparing the integration between ABC4EU Touchpoint solutions (ABC systems, Mobile Stations and Enrolment Stations) to be in compliance with the new features that were not included in the first version of EES, BMS NFP and TDMS/CES subsystems.
The integration activities ended with the preparation/deployment of the Pilot Solutions (touchpoint and backend systems installation).
With the change of EES to a traveller account based system rather than a travel document based system this meant a complete set of new EES Services that had to be integrated and debugged through testing to the Front End Gateway (FEGW) and the Enrolment Station and Mobile System (end clients). Additionally new services were made available for biometric face matching in EES/BMS for both TCN VE & VHs as part of the biometric fusion capability integrated on the Mobile System, and biometric duplicate search on the Enrolment Station integrated with EES/BMS.
WP6 BCP Test & Validation
Summary of the tasks carried out and significant results
PwC effort was focused on continue with the validation activities for the second round of pilots and the elaboration of related reports.
The checklists of the validation activities that were elaborated internally by PwC (based on the information included in deliverables 6.2 6.3 and 6.4) were updated once the results collected from Madrid and Lisbon (first round) pilots were analysed.
Likewise, PwC updated the set of forms elaborated to facilitate the collection of data splitting the uses cases in smaller tests. For this purpose, in addition to the forms elaborated, the observation was a key method to collect these results properly along the first PoC.
The validation results and the tests performed for the systems involved in the first PoC include:
- User and ergonomic tests
- Functional tests
- Mobile integrated solution tests
- System security tests
- Data protection tests
In addition, PwC was responsible for analysing the feedback collected by Eticas from the travellers that agreed to participate along the first PoC. Likewise, PwC gathered the border guards’ feedback from the different locations (Madrid, Lisbon and Algeciras).
Potential Impact:
Potential impact
It has been identified since the preparation of the proposal that the change in the regulation regarding border management, and the development and integration of a central, EU-wide, Entry Exit System, and a national facilitation program, will imply a huge change in border activities in terms of human resources, time required for border clearance, annoyance to the travellers (as additional processes have to be carried out, prior or during the border checks) and footprint for new equipment at the border crossing points, among others.
One of the main objectives of the project was to assess this impact by the deployment of functional systems covering the technical scope of the proposed solutions, under real operational scenarios, in order to identify further ways of automation that facilitate and alleviate the negative impact of its deployment.
As it has been observed during conversations and surveys with the travellers during the pilot execution, extending the automation to third country nationals is very welcome by the travellers, as the usually feel that TCNs are discriminated in some way against EU citizens because they can’t avoid waiting at the border queues.
Systems such as those developed in ABC4EU project, deployed at the borders integrated with the Entry Exit System and National Facilitation Program will help ensuring a seamless border crossing for third country nationals while keeping the security levels at the borders.
Overview of dissemination activities and exploitation of results
Overview of main dissemination activities & channels.
In accordance with the DoW, the following major dissemination activities were performed within the entire project’s duration and in part they will continue after the project has ended:
Official Logo and Official ABC4EU project website.
Dissemination Materials, such as hard copies of Leaflets, posters, flyers and roll-ups created and distributed in various events such as pilots, conferences, workshops and so forth.
Marketing Material such as: Shoe bags, luggage tags, and small bottles for storing liquids for travellers, all bearing the ABC4EU project logo, as a way to promote the project have been distributed in conferences, abc4eu final conference, end user workshops and project pilots.
Social media promotions of project milestones, events, or other information worth sharing with the public audience through project Website, Facebook, Twitter, LinkedIn and YouTube.
Forum
Forum has been created in order to promote discussions from a wider audience regarding privacy, technology used and legal aspects of ABC gates. Link to the forum: www.forum.abc4eu.com However due to lack of interest it has been discontinued half way through the project.
Twitter is a social media platform that is used to quickly share short status or news reports and/or refer traffic to the main project website. Twitter has been chosen for its short and quick communication abilities without giving too much information away. Also another reason is that twitter is used actively by most of our partners and end users.
Twitter name: @ABC4EU
Facebook is a social media that has been primarily used to share public deliverables and make any announcements necessary related to the project. For example newsletters, press releases, scientific publications, presentations and other dissemination activities performed by ABC4EU partners have been shared there. The project link is: https://www.facebook.com/abc4eu
LinkedIn is a professional networking platform. Laurea has created an ABC4EU-project group to which all news has been shared. LinkedIn has been used to promote the seminar to expand the end-user community. Also, external links, photos and such material. LinkedIn name: ABC4EU-project
YouTube channel:
Several videos showcasing the ABC4EU project technology, process flow, and other project related information have been produced and uploaded to social media and project YouTube channel. These have also been published in the project website.
.
Newsletters: Considering the newsletter definition, it is a regularly distributed publication about a particular topic which is of specific interest to receivers. Thus in context of ABC4EU project, thematic newsletter has been distributed regularly in order to notify relevant parties regarding the project progress. These were also published in project website, social media and sent to project subscribers.
Press releases: relevant media channels have been identified and then selected according to their best suitability to the project needs. In the case of ABC4EU project all varieties of published short releases have been named “press release” independent of the media type. At least seven press releases have been released during the project.
Events: The ABC4EU project has been effectively presented within relevant conferences, seminars, workshops, fairs which have been thematically related to the project’s scope. The main goal of this participation was to raise awareness of the project among different target groups to the best of abilities. Within such events ABC4EU project has been promoted. The presentations have been concentrated on the area of interest of event participants. The project has been present both in research conferences and national events addressed to practitioners I.e. Students, public administration employees, private enterprise employees, academic institutions representatives and so forth. Events such as keynote speeches, paper publications, conferences, presentations, lectures and fairs, have been conducted periodically whenever possible such as the SRIEE 2017 event and the Nordic Business forum.
Pilots such as dissemination of the project via direct contact with various end users such as airport operators, border guards and travellers. In These events all project promotion materials have been distributed such as leaflets, flyers, roll-ups especially created for the pilots, shoe bags, liquid bottles and luggage tags.
Overview of main exploitation activities
The main exploitable outputs have been conducted via
Development and improvement of dissemination strategies, through lessons learned methodology for better future dissemination outputs.
Technology development whose applicability is not only limited to ABC gates but to a wider usage area, such as algorithms for abandoned objects for increased security, look&walk for two step segregated topology and so forth.
ABC4EU consortium have promoted the project results through papers, articles and publications on specialised newspapers and journals.
The objective has been on one hand to obtain a great visibility among research communities and target groups (avoiding wasting resources disseminating the project through non-specialised press communications), and on the other hand, it has been to inform and reach research communities and experts in the targeted sectors through scientific journals. All published articles are reported in the website section “publications”. Also after the project has ended statistical information from traveller interviews and workshops will be used for publishing new papers regarding ABC gates and the EES.
Skills development, training for new staff at airports, mobile system development SDK training for employees, Teaching I.e. integration of the project with the study units at universities, incorporation of the acquired knowledge in educational activities, such as BA, MS and PhD programs and summer schools.
Learning from experience such as: Applying lessons learned methodology and carrying out detailed analysis in order to take advantage of the conclusions and results obtained in order to address the latest tests planned in the project in the adequate manner.
Contacts with other FP7 / Horizon2020 –projects
Project partners created and maintained contacts with other FP 7- projects in order to share experiences. Joint activities included but were not limited to:
• Cross-referencing other projects’ websites. In this way the visitors of a project website could be easily directed to other similar projects funded by FP 7 and find out about projects they were not previously aware of.
• Distributing promotional material to other projects events. Through such cooperation, distribution of leaflets and promotion of the project is easily multiplied with minimum effort and without any additional costs. Moreover, it provides the opportunity to disseminate to events that none of the partners is present and to countries outside the consortium.
• Joint participation at seminars and conferences.
Cooperation with other similar projects together with end users from academia, legislation, border security and ethical experts, in order to share information, learn from one another and discuss project results and expectations from end users, legislation, ethics and technological perspectives.
Dissemination materials
Project Logo
The project logo was developed in the beginning of the project. The general shape was agreed by the project partners that the logo shall refer to the main issue of the project. The main goal of the project logo is to create awareness of the ABC4EU and attract the attention of potential target groups. The logo is clear and simple and may be easily recognised and relevantly associated. It contains the abbreviation of the project which includes ABC4EU name and the e-Passport Chip.
The logo has been provided in different formats and resolutions to be comfortably suitable for different purposes i.e. document, website, print and so on.
Website
The ABC4EU website (www.abc4eu.com) has been the main part of the project dissemination. The website consists of two areas: the public and private. From the dissemination’s perspective the most important part is the public side of the website which is available for the general public and will contain information related to the project e.g. a project summary, news, information on the consortium, public publications and so forth.
The project website is mainly utilized to raise the awareness among target groups’ representatives as well as to be a source of information related to events, publications and so on.
There are three main goals for the website:
• Ability to provide detailed information concerning the project aims, progress and news
• Hold publications in disposal for any interested party
• Collect internal documents and ensure their availability among the project partners.
The website has been regularly updated, news has been introduced, publications have been published, and the project’s progress has been generally reported.
Poster
The poster has been adjusted to the current project’s stage and has been promoting current project’s achievements, it has also been adjusted to the target audience in case of preparations for a previously planned larger event.
Poster’s aim is:
• To inform on current project’s stage and achievements;
• To attract stakeholders and draw their attention to the ABC4EU project, its goals, outcomes and achievements;
• To share only the key issues of the project and to encourage people in finding more information (e.g. visit project's website, attend the course or any event where the project will be presented.
The first project’s poster, prepared at the very beginning phase of the project was concentrated on the general project’s purpose. Its aim was to announce the project, its goals and planned outcomes. It is addressed to the general public and to researchers.
The poster schema is consistent with the project logo and leaflet colours and character. The picture represents the initial poster’s appearance and content.
Roll-Up
Laurea UAS has also developed roll-ups according to the uniform design of all ABC4EU dissemination material. This material may also be printed by partners and used in events where ABC4EU is presented. The Roll-up has been created with the aim of attracting potential interested parties in events such as fairs, project presentations, lobbies on meeting venues, and also have been used in pilots for providing more information to travellers as well as to attract their attention.
The first roll-up has been created on M9 as an addition to the overall dissemination material. This type of material shall include the project logo, marketing phrases and funder information.
Leaflet
The ABC4EU leaflet has been adjusted several times to the current project’s stage and promotes current project’s achievements. Throughout the project there has been also possibly to adjust the leaflet to the target audience in case of preparations for a previously planned larger event.
The first project leaflet has been prepared at the very beginning phase of the project and is concentrated on the general purpose of the project. Its aim is to announce the project, its goals and planned outcomes. It is addressed to the general public and to researchers.
Towards the end of the project, particular outputs have been described in the leaflet, in a timely manner.
Flyer
A flyer has also been added to the overall dissemination material by the M9. Similar to the roll-up, the flyer is designed as a quick touch point to the project. It consists minimal information about the project including: project logo, funding, marketing phrase as well as addresses to the online presence of the project. The leaflet has been designed to attract the attention of readers and increase their interest towards receiving more information regarding the project.
Marketing Materials
Marketing materials have been shared in various events where ABC4EU project has participated, especially in the pilots, where materials were given to travellers after they had accepted
List of Websites:
www.abc4eu.eu
Twitter name: @ABC4EU
Facebook: https://www.facebook.com/abc4eu
