Detection and Analysis of Graphic Images
LETI - Detection of Graphic Images (Various Format Graphic Images) by Analyzing the Given Fragment Among Number of Various Computer Data
Tech Area / Field
- INF-SOF/Software/Information and Communications
- INF-DAT/Data Storage and Peripherals/Information and Communications
- INF-IMA/High-Definition Imaging and Displays/Information and Communications
3 Approved without Funding
Russian Academy of Sciences / Institute of Applied Physics, Russia, N. Novgorod reg., N. Novgorod
- Raduga-soft, Ltd, Russia, N. Novgorod reg., Sarov\nССT – Technology of Chaos, Ltd, Russia, Moscow\nForensic Science Center of the Ministry of Home Affairs of the Russian Federation, Russia, Moscow
- Excom Inc., USA, NJ, Old Bridge\nSymmetrics Gmbh, Germany, Hannover
Project summaryThe objective of the project is to carry out applied research, development, testing and demonstration of a system model comprising a complex of software modules, which allows analyzing different types and formats of graphic images among great number of various computer data. Using the “Model of Software System for Analysis of Graphic Images” (MSSAGI) is focused on increasing the efficiency of the work being accomplished by the employees of the Forensic-Science Center of the Ministry of Interior of Russia (FSC MIR).
The system to be made will enable the following operations:
1.1. viewing a great number of graphical raster formats and recognizing graphics files by signatures and file structure, irrespective of extension of those files, by creating the lists of found file groups: only graphics, only non-graphics, possibly graphics (for further heuristic analysis); analysis of vector files;
1.2. recognizing and viewing text files containing graphic images;
1.3. searching graphic images basing on image pattern given by expert;
1.4. searching graphic images containing text basing on text fragment given by expert;
1.5. – making hint lists of found images that can be sorted by different criteria (e.g. degree of similarity, dimensions, dates, extension, image format, etc.).
The system efficiency will be tested at:
- comparing test graphic images (approved by FSC MIR) by degree of similarity;
- searching obscene and obscene pictures;
- searching the images of banknotes or their fragments.
It is anticipated that the project activities will consist of two stages.
The first stage (1 year) implies that, basing on the available techniques, the software for modules of search and identification of images of a limited set of graphical formats (e.g. bmp, jpeg, gif or other available formats) will be developed.
The second stage (2 year) includes testing of created software and enhancing its functionalities for other types of graphic images the structure of which will be realized by developers at the first stage of work. A system model with matched interfaces and formats of data exchange between developed modules will be implemented.
As a result, the software modules of the system will run with selectable format of images and converters will demonstrate possibilities of handling the files of the most of existing graphical formats and presenting them in the chosen format for system images.
The software modules of the system will provide an option of viewing the images that are created and processed by most of known graphics packages (built in OS WINDOWS, MS Word, ADOBE ACROBAT, DjVu,). In the progress of work the list of processed file types of other packages will be extended.
Also, the software modules of the system will provide an option of searching image files of the following graphical formats: BMP, PNG, GIF, PCX, DCX, TIFF, JPEG, WMF, EMF, PDF, DjVu, Targa. In the course of the project the additional lists of graphical formats will be specified which will be supplemented to the system at the first and second work stages.
The software system to be created will enable search of the files containing only graphic images as well as the files in which a graphic image is part of file, e.g. MICROSOFT Word, pdf:
The system will run under Windows XP and perform file control under OS Windows 2000 and higher (except for Vista).
The system does not imply operation with protected data. The system will be able to detect such data (files) but not crack them.
Technical Approach and Methodology
Structurally, the system includes the following program suite:
- Data pre-processing programs.
- Programs enabling search of all files in the system which are suitable for processing.
- Viewer for found files.
- File processing scheduler.
- File handler.
- File processing modules. All file processing modules in the system are subpided into filter and indexing ones.
- Filter modules:
- Search of graphics files with banknote images
- Search of graphics files with obscene images
- Search of text and graphics files with images or templates of documents e.g. passports, vehicle certificates, state customs declaration, etc.
- Search and analysis of text in graphics files
- Indexing modules:
- Creation of full-text index for handling full-text requests basing on text materials from processed information carrier.
- Creation of graphics file database for searching images similar to those given by expert.
The project participants are co-operating specialized groups, each having long-term experience in developing software systems for processing, analysis and encoding of text and graphic information. The expertise of the project involved groups is illustrated by the list of 20 publications /1-20/.
Expected Results and their Application
The project proposed involves the following R&D categories: applied research, development and demonstration.
The expected project results are:
- Operating MSSAGI model.
- Installation programs, operation documentation that can be offered for use in other expertise and criminalistics departments of RF MIR.
- Developed algorithms and codes of inpidual modules that can be commercialized, even separately from MSSAGI.
- Technical basis in the form of MSSAGI software implementation and programming experience for developing similar applied techniques on customers’ demand.
- Publications and advertising materials.
Fields of Application:
The system can be used in any organization making computer expert examinations (on the basis of the number of experts).
Development of commercial program for search and filtration of Internet Web sites with obscene and obscene content.
Development of “photo-archive” commercial program with photo-image indexation according to specifics of selected image fragments (e.g. people faces on photos).
It is anticipated that collaborators of the project (Excom.Inc, USA and Simmetrics, Germany) will participate in commercialization of the project results and their promotion to the western markets.
The Project proposed meets the ISTC Goals since the present development work:
- Serves only peaceful purposes;
- Project result possesses commercial potential;
- Directly employs the scientific potential of former RFNC-VNIIEF labs specialists as well as FSC MIR RF and IAP RAS employees who have been earlier involved in nuclear and other scientific military programs, which will guarantee their alternative employment and integration into the international scientific community.
The International Science and Technology Center (ISTC) is an intergovernmental organization connecting scientists from Kazakhstan, Armenia, Tajikistan, Kyrgyzstan, and Georgia with their peers and research organizations in the EU, Japan, Republic of Korea, Norway and the United States.
ISTC facilitates international science projects and assists the global scientific and business community to source and engage with CIS and Georgian institutes that develop or possess an excellence of scientific know-how.