Objective
The project aims to develop technical solutions for the implementation of a tool for historical research. The objective is to design and realise a prototype image-data bank of manuscripts related to Venetian heraldry, and provide the tools for intelligent searching and retrieval of the images.
The latest scanning and imaging techniques will be used together with intelligent hyper imaging techniques to provide a tool that can be used by scholars to research the old manuscripts without having to access the original books. Among the innovative aspects of this project is the production of a consultation tool able to retrieve images and associated information from the data base, following intelligent browsing paths which will be constructed dynamically, based on text and image associations. Users will be able to search the image-base and establish paths and links between images and texts by querying using both a multilingual interface and also a pattern-recognition optical interface (which will be in the case of an enquiry based on optical information such as an armorial bearing).
To test the effectiveness of the tool the target domain application will include the books of arms, armorial bearings and genealogies of Venetian families. The approach can be easily extended to many other contexts where images of the original manuscripts are needed for research.
Impact and expected results:
It is expected that the image database and its integrated toolset, a complete research tool for heraldic data, will dramatically enhance the service given by the library to its customers, through its advanced image retrieval and hyperlinking of information. Furthermore, copies of the image database could be used by other libraries and historians without travel to the originating library and without risk to original documentation.
The service ought to be cost-effective as duration of research is likely to be reduced from weeks to minutes and services that were impossible will be able to be offered (e.g. specific searches and the distribution of high resolution reproductions for further study). Updating and incorporating new material will be automatic.
Deliverables
The main deliverable will be the software tool for accessing of manuscripts. Design, development and test documentation - and the exploitation plan - will be confidential or restricted category.
In the public domain will be:
User requirements and specification report;
Technical reports on the alpha and beta test versions;
Report on transferability testing;
Final report and demonstration.
Technical approach:
The project was organised into six workpackages:
User requirements and specifications which defined the technology and testing to be employed, tests on photographic and scanning techniques and standards, and preliminary design;
Image and text database creation with selection of texts plus photography and scanning of images to resolution and quality standards;
Development of alpha version - notably the intelligent browser;
Testing of system components and final review of the specification on-site with library staff, and including data transferability and refining of the image database;
Development of beta version for final system development;
Final testing and report on-site, with technical and transferability tests, and user evaluation.
The images will be stored in both high (up to 4000 dpi, for reproduction purposes) and low (for retrieval purposes) resolution. A cylindrical scanner will be used to produce the high resolution images from the photographs taken of the manuscripts. A user-friendly interface will be developed. The system will be generic and thus reusable for other related image pattern recognition systems. It will also be portable between DOS and UNIX systems. Relevant standards such as SGML, HyTime, TIFF and high definition JPEG will be taken into account. Fuzzy image search and Artificial Neural Network techniques will be tested.
Key issues:
- User requirements and scanning techniques;
- Standards for image storage and retrieval;
- Photography and scanning;
- Design and implementation of an intelligent browser;
- Testing of transferability to different sites and collections;
- Exploitation plan.
Fields of science
- natural sciencescomputer and information sciencessoftware
- natural sciencescomputer and information sciencesdatabases
- natural sciencescomputer and information sciencesartificial intelligencepattern recognition
- natural sciencescomputer and information sciencesartificial intelligencecomputational intelligence
Topic(s)
Call for proposal
Data not availableFunding Scheme
Data not availableCoordinator
30124 Venezia
Italy