Head of Sector – Dr. Nikolay Merzlyakov

The leading researchers of the laboratory include:

• development of relative databases and DBMS;
• optical-digital methods of image processing and pattern recognition;
• synthesis of two-dimensional digital filters;
• digital image enhancement;
• loss and lossless compression of gray-scale and color images;
• colorimetry;
• multimedia;
• digital holography;
• three-dimensional scene reconstruction;
• medical image analysis and classification.

Methods and tools for digital processing of watermark images and relational watermark database were developed and elaborated. All proposed methods and tools were integrated in a system. This approach gives all needed possibilities to perform watermark classification by means of the system.

A relational database of the Russian Academy of Sciences was created. It includes biographic data and digital photo-archive of all members of the Academy during its 275 years history. Special software tools were developed and elaborated for database management and data usage. The database and software were published as the CD-ROM disk “The Russian Academy of Sciences. 1724-1999”. This disk is a unique electronic publication. The data volume is 650 MB.

Local adaptive signal processing on the base of short-time discrete cosine transform (DCT) was proposed. Two fast recursive algorithms for computing the short-time DCT were suggested. The algorithms are based on a recursive relationship between three subsequent local DCT spectra. The algorithms require essentially less operations of multiplication and addition comparing with known fast DCT algorithms. The inverse transform for signal processing in a running window was also proposed.

A new noise suppression technique on the base of short-time discrete cosine transform was proposed. Computationally efficient algorithms were suggested and implemented for real-time speech processing in actual noise environment such as the helicopter cockpit.

A new algorithm to compute precise 3D shape of a moving object for color motion stereo was proposed. Input data was obtained from a single color CCD camera and a moving belt. 3D-shape recovery in motion stereo was formulated as a matching optimization problem of multiple color stereo images. It was shown that the problem of matching among multiple color motion stereo images can be effectively carried out by use in of the circular decorrelation over a color signal. 3D shape recovery using real color motion stereo images demonstrated a good performance of the algorithm in terms of reconstruction accuracy.

New methods were developed to support the analysis of complex scenes. The expert’s domain knowledge was used to develop strategies for image processing, analysis, and interpretation. That helps to look for specific features in an image. Effective rules (models of decision-making) for image classification on the base of observed features were proposed. The methods were tested in the task of early differential diagnosis of solitary lung nodules, and they helped to improve accuracy and reliability of peripheral lung cancer diagnostics.

Object-oriented analysis, design, programming, and debugging to automate and simplify detailed steps of protocol for the scanner console of Hitachi Computer Tomograph were performed.

A software for CT-guided surgical procedure was developed. The designed software module checks out the right way based on CT images and provides an optimal navigation of surgical instrument inserted into the patient body.

• Austrian Science Fund FWF (FWF-13289-ARS): “Watermarks in Middle Age Manuscripts”.
• Russian Foundation for Basic Researches (No. 99-07-90017): “Development of both watermark image database and system for watermark image processing and analysis”.
• The project of the State scientific and technical subprogram “Information of Russia” of priority direction of “Information technologies and electronics” (No. 037.03.319.17/1-99): “Development of integrated system for storage and processing multidimensional signals for information maintenance of scientific researches, culture and education”, subpart "Development of technology and software for construction of multilevel text-graphic digital archives on the basis of conventional database management system”.
• Russian Foundation for Basic Researches (No. 99-01-00265): “Development of algorithmic and physical model of 3D view device”.
• Russian Foundation for Basic Researches (No. 99-01-00269): “Development of optical-digital methods for real-time pattern recognition”.
• Russian Foundation for Basic Researches (No. 99-01-00268): “Development of methods to acquire knowledge for image interpretation“.
• INTAS (No. 96-0833): “Development and investigation of fast recursive algorithms for image processing and data reduction using segmentation”.

1. Belikova T.P., Stenina I.I., Yashunskaya N.I. An approach to robotics diagnosis of medical images on the base of image processing and syndrome features analysis. Proc. of the 51th ICB seminar on medical robotics, biomechanics and musculoskeletal system, Poland 1999, J. Medical Robotics, 1999, Med. Robotics (in print).
2. Belikova T.P., Stenina I.I., Yashunskaya N.I. Image interpretation support by means of image processing and syndrome feature analysis. Proc. of the Conference “Medical Cybernetics in Medical Practice”, Moscow, GVKG Burdenko 1999, pp. 115-118.
3. Belikova T. New informative technologies in Medicine. European school of oncology. Russia, Moscow, June 1-3, 1999. 12 p. (in Russian).
4. Belikova T.P., Stenina I.I. Computer-assistance methods to improve image analysis and interpretation in the case of uncertainty. SPIE Symposium on Medical Imaging (accepted).
5. Wenger E., Karnaukhov V.N., Merzlyakov N.S., Haidinger A. Digital image processing and database of medieval watermarks. Proceedings: New Information Technologies in the Cultural Area in new millennium. Center PIC, STG, 1999, pp. 3/2/1-3/2/4.
6. Wenger E., Karnaukhov V.N., Haidinger A. Contour extraction of watermarks in old manuscripts. In: Eurographics’99, short papers, (Maria AlbertaAlberti, Giovanni Gatto and Ivan Jelinek, editors), Milan, 1999, pp. 236-238.
7. Karnaukhov V.N., Kuznetsov N.A., Merzlyakov N.S., Rubanov L.I. Text-and-Graphics Digital Archive of Russian Academy of Sciences. Proceedings: New Information Technologies in the Cultural Area in new millennium. Center PIC, STG, 1999, pp. 5/2/1-5/2/5.
8. Karnaukhov V.N., Kuznetsov N.A., Merzlyakov N.S., and Rubanov L.I., et al. CD-ROM disk “The Russian Academy of Sciences. 1724-1999”, IITP RAS, Moscow, 1999.
9. Karnaukhov V.N., Merzlyakov N.S., Rubanov L.I. Image processing and storage in digital archives of manuscripts and photo-documents. Infix Publ. (Germany; 1999, in print).
10. Karnaukhov V.N., Merzlyakov N.S., Wenger E., Haidinger A. Digital methods of image processing and identification of medieval watermarks. Computer Optics, Moscow, 1999 (in print).
11. Karnaukhov V.N., Kuznetsov N.A., Merzlyakov N.S., Rubanov L.I. Design principles of Digital Archive of Russian Academy of Sciences. Computer Optics, Moscow, 1999 (in print).
12. Kober V., Choi T.S., Ovseevich I.A. Multichannel pattern recognition based on projection preprocessing. Pattern Recognition and Image Analysis, Vol. 9, No. 3, 1999, pp. 437-447.
13. Kober V., Ovseevich I.A. Use of information redundancy of signals for reducing the computational costs of signal processing. Radiotekhnika, No 5, 1999, pp. 13-16. (in Russian, translated in USA by Allerton Press Inc. as Radio and Communication Technology).
14. Kober V., Cristobal G. Fast recursive algorithms for the short-time discrete cosine transform. Electronics Letters, Vol. 35, No. 15, 1999, pp. 1236-1238.
15. Kober V., Cristobal G. Rank order filtering based on accelerated algorithm of local histogram calculation. SPIE Annual Meeting, Applications of Digital Image Processing XXII, July 23, 1999.
16. Kober V., Seong Y.K., Choi T.S., Ovseevich I.A. Trade-off filters for optical pattern recognition with nonoverlapping target and scene noise. Pattern Recognition and Image Analysis (in print).
17. Kober V., Choi T.S. Single-output multichannel pattern recognition with projection preprocessing. Optical Engineering (accepted).
18. Mozerov M., Kober V., Choi T.S. Color motion stereo based on adaptive correlation matching. SPIE Annual Meeting, Applications of Digital Image Processing XXII, Denver, 1999.
19. Mozerov M., Kober V., Ovseevich I.A., Choi T.S. Motion stereo matching using a modified dynamic programming. Pattern Recognition and Image Analysis (in print).
20. Nelson D., Cristobal G., Kober V., Cakrak F., Loughlin P., Cohen L. Denoising using Time-Frequency and image processing methods. SPIE Annual Meeting, Applications of Digital Image Processing, 1999.
21. Park M.S., Mozerov M., Kim D.Y., Roh K.S., Choi T.S. Object shape recovery in lateral navigation system using motion stereo technique. Proc. IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems, Taipei, Taiwan, August 1999, pp. 273-278.
22. Perner P., Belikova T. Data Mining in Picture Archiving Systems. 3rd Intern. Conference on Practical Application of Knowledge Discovery and Data Mining. Proc. (PADD-99), ISBN 1 902 426 045, 1999, pp. 207-210.