Cosmic Ray Tomograph for Identification of Hazardous and Illegal Goods hidden in Trucks and Sea Containers

Project Information

SilentBorder

Grant agreement ID: 101021812

DOI

10.3030/101021812

Project closed

EC signature date 27 April 2021

Start date 1 May 2021

End date 31 July 2025

Funded under

• Secure societies - Protecting freedom and security of Europe and its citizens

Total cost

€ 7 471 536,25

EU contribution

€ 7 471 536,25

Coordinated by

TARTU ULIKOOL
Estonia

Periodic Reporting for period 3 - SilentBorder (Cosmic Ray Tomograph for Identification of Hazardous and Illegal Goods hidden in Trucks and Sea Containers)

Reporting period: 2023-11-01 to 2025-07-31

Summary of the context and overall objectives of the project

Illegal goods are trafficked into Europe by criminal organisations using a range of methods and tools which are very diverse and adaptable to specific border conditions. In 2018, EU customs seized 472 tonnes of drugs, 4.1 billion illegal cigarettes, 2600 firearms, over 3.6 million pieces of ammunition and 476,00 pieces of explosives. Yet it is estimated that only 5-10% of illegal goods are even detected with conventional x-ray scanning technologies.
Today, the most widely used technology for scanning vehicles is gamma-ray and x-ray radiography. Both have some technological shortcomings. The scanned images are incomplete in terms of detecting the materials and identifying contraband depends on the operator’s subjective opinion. The clearance time is long and detection labour intensive. The equipment is costly and inconvenient for end-users because of the safety regulations and requirements that need to be followed. The technologies typically make use of artificial ionizing radiation which is harmful to humans and animals.
The main objective of SilentBorder was to develop and validate a new cutting-edge technology cosmic-ray tomography (CRT) scanner for border guard, customs, and law enforcement authorities. The CRT scanner is safer and easier to deploy as it is based on natural cosmic rays and does not produce harmful radiation. Contrary to conventional x-ray or gamma-based imaging techniques, the SilentBorder scanner solution is able to distinguish between different materials and localise them inside the cargo or vehicle by providing visualised 3D images. The technology uses machine learning algorithms which allow for automated and remote management with less human interaction, meaning there is also less risk of corruption and/or human error among border workers. Our aim in the future is to offer faster clearance times compared to industry standard x-ray systems and also have low maintenance costs for the final product.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

The SilentBorder project has advanced the development of cosmic ray tomography as a powerful complementary tool to X-ray scanning for detecting hazardous and illegal goods in trucks and sea containers.
A major technical achievement was the design and validation of the hodoscope module, the core detector element comprising muon-sensitive scintillation fibers. A prototype was built and tested successfully, proving the feasibility of the production technology.
Alongside this, the Data Acquisition System (DAQ) was developed and tested. This is especially complex due to the large number of fibers in each hodoscope.Future R&D for such large-channel systems was mapped out.
DAQ development was accompanied by integration of a special nuclear materials detection solution - the SNIPER - into the cosmic-ray tomography technology.
To guide system optimization, a comprehensive Monte Carlo modelling framework was established to simulate muon transport and detection in a realistic muon tomography system. A detailed computer model was developed to describe the muon tomography system and different cargoes, evaluate the performance of the cosmic-ray tomograph and suggest ways to optimize detectors and analysis methods. Many scenarios have been simulated, the data and images stored for future use as templates for material identification. A new code TomOpt was developed, based on simplified simulations and a novel programming technique (differentiable programming), to optimize muon tomography system designs. An assessment of the parameters of the designed muon tomography system using simulated data revealed a few improvements that could enhance its performance and lower costs. These improvements may require further development of detectors and analysis methods, in particular, using machine learning techniques such as neural networks.
A core part of the SilentBorder project was the development and optimization of the tomography system as a whole. A physical tomography system was designed, incorporating algorithms that facilitate tomographic measurements. Key achievements include the development and initial validation of tomographic reconstruction methods using machine learning, and classification methods that have demonstrated over 90% accuracy with simulated data for simple shape detection and material classification. Technical design documentation and drawings, as well as a report on the system's relocatability and flexibility were produced. Software for detector tracking, alignment, and data analysis was also developed.
The SilentBorder system was tested and demonstrated to the public on three separate occasions.
The 1st demonstration focused on providing the basic functionality of the system. Small scale prototype successfully detected muons and produced images. The early version also integrated AI algorithms to help interpret the data, demonstrating how artificial intelligence can enhance imaging even in the project's initial stages.
In the second demonstration, the SilentBorder technology was shown to be fully compatible with the SNIPER special nuclear material detection system developed by CAEN. This highlighted the system's flexibility and its potential to work alongside existing nuclear security tools.
The 3rd and final demonstration took place in a realistic outdoor environment. A fullscale prototype was tested together with X-ray scanning technology. By innovative AI methods, the system successfully identified materials missed by X-rays alone in under 50 minutes.
Together, these results demonstrate SilentBorder’s strong potential to provide a safe, effective, and non-invasive inspection method for border security. By combining cosmic ray tomography with advanced AI-driven analysis, the project has opened the way towards future systems that can reliably detect concealed materials and support customs and law enforcement agencies in protecting society. In addition to advancements in identifying low atomic mass materials, the integration of the patented SNIPER system means that SilentBorder will also push boundaries in the detection and identification of special nuclear materials.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

SilentBorder has provided a technology that can be used as a complementary method to X-ray scanning with key advantages: no additional ionizing radiation, discrimination of materials based on density and composition, and delivery of 3D data.
The approach relies on cosmic-ray tomography, combining muon scattering and transmission, and secondary particle analysis. As a result, the technology enables identification of shielded materials and, through advanced data processing, allows material discrimination and anomaly detection. SilentBorder has addressed multiple challenges in developing cosmic ray technology. For light materials, scattering effects are small and differences subtle, yet the project has shown potential to discriminate even between organic substances.
Imaging algorithms have already benefited from new concepts introduced by the project. SilentBorder has outlined pathways toward commercially viable hardware solutions.
Scanning times, another potential limitation, have been optimized through identified use-cases and strategies to reduce duration. Before the launch of the SilentBorder, engaging end-users was challenging due to the low technological maturity of the technology. SilentBorder has now created opportunities to involve end-users in the development of cosmic ray tomography for detecting illegal goods.

SilentBorder Logo