INSTITUTE OF INFORMATION TECHNOLOGIES - BAS

Department: Pattern Recognition and Biometrics
Head of Department: Associate Professor Georgi Gluhtchev, PhD

Current activities of the PR&B department are concentrated on R&D in the areas of (1) Image, Video and Speech Analysis and Recognition, (2) Content Based Image/Object Retrieval (CBIR/CBOR), (3) Biometric based identification, as well as other areas like Neural Networks for Classification, Computer assisted Medical Diagnostics, etc.
The Department has long-term interests in the technological development of information systems using Image Processing and Recognition.
The Department of Pattern Recognition & Biometrics consisting of: The Department actively participates in European and National R&D projects and presents the results in international and national journals and conferences.
The Department is co-organizer of the annual international conference CompSysTech (2000–2008).

Participation in European R&D Projects:
FIVES (2009-2010): Forensic Image and Video Examination Support, “Safer Internet Plus” Program of EC, SIP-2008-TP-131801. Consortium - Karlstad University (coordinator), NetClean Tech. AB (Sweden), IIT-BAS (Bulgaria), DFKI (Germany), KLPD (Netherlands) and FCCU (Belgium) The objective of the FIVES project is to develop novel investigative tools specifically tailored for investigations involving images and videos of child sexual abuse. The tools will be based on already existing research and software created by the academic partners that will be adapted, extended and integrated into an easy to use tool set that fits the police requirements. The expected end result of the FIVES project is a self-contained tool set distribution that is easy to use and that supports easy translation to languages across Europe.
ETN TRICE (2008-2010): Teaching, Research, and Innovation in Computing Education, Erasmus Thematic Network No. 142399-LLP-1-2008-1-BG-ERASMUS-ENW. Consortium – 70 partners from 30 European countries, coordinator - the University of Rousse, (http://trice.ecs.ru.acad.bg), IIT-BAS partnership The project is designed to promote research and innovation in Computing education practice. The targeted objective will be achieved through analysis of the state-of-the-art, exchange of positive experience, dissemination of the good practices between the involved partners, improvement of the quality of teaching through employment of innovative educational technologies. The project goal also includes the integration between universities, research institutes and companies, so that the ETN-TRICE network grows to the level of operator of the teaching, research and innovations in the area of Computing Education.
COST 2101 (2006-2009): Biometrics for Identity Documents and Smart Cards, Consortium of 15 European partners, coordinated by the Swiss Federal Institute of Technology. The goal of the project is to stimulate the cooperation of European researchers in the area of document and smart cards identity. Different biometric modalities are involved and their potential applicability will be evaluated.
BIOSECURE (2004-2007): Biometrics for Secure Authentication, Network of Excellence: 30 European partners and two USA companies, coordinator: Groupe des Ecoles des Telecommunications in Paris.
Contract: 507634-IST-FP6-ÅÑ, (http://biosecure.it-sudparis.eu/AB/), IIT-BAS partnership
The main objective of this network is to strengthen and to integrate multidisciplinary research efforts in order to investigate biometric-based identity authentication methods for the purpose of meeting the trust and security requirements in our progressing digital information society.
ETN DEC (2004-2007): Doctoral Education in Computing, Erasmus Thematic Network No.114046-CP-1-2004-1-BG-ERASMUS-TN, Consortium – 70 partners from 30 European countries, coordinator - the University of Rousse, IIT-BAS partnership (http://ecet.ecs.ru.acad.bg/etndec) The main project objective was to help the process of affirming the PhD education as “the third” level of the higher education, to reveal the possibilities to increase the quality of doctorate education, to establish methods and instruments for the estimation, approbation and certification of the knowledge and skills of the PhD students in the area of Computing, that would be synchronized to the practices in Europe and around the world.
Greek-Bulgarian Collaboration Project on "Study of Biomedical Data by the Methods of Multiresolution Analysis. Polyspline Wavelet Analysis in Immuno-Computations and Brain Research". Research project B-G-17/2005, co-funded by the National Science Fund of Bulgaria, BG team coordination by IMI- BAS, IIT-BAS partnership (2005-2007) The main purpose of the Project was to unify the efforts of the specialists from two different research teams in Greece and Bulgaria, in the interdisciplinary area of Biomedical Informatics and more precisely, in the applications of modern mathematical methods to the analysis and approximation of biomedical data. The team from Bulgaria has been more concentrated over the analysis of the EEG data, while the team from Greece – over the DNA data.
Requirements and Framework for Environment and Transport Telematics, (1998-1999): CAPE Project (TR 4101/ IN 4101), FP5 of EC, IIT-BAS partnership The aim of the Bulgarian team in the project was the preparation of a regional report for the country’s status in the area of environmental and transport telematics.

Participation in National R&D Projects:
ASTROINFORMATICS (2009-2011): Signal Processing and Analysis of Digitized Astronomical Data and WEB-Based Implementation. Research project DO-02-275/2009, funded by the National Science Fund of Bulgaria, Consortium of 4 institutes of BAS (IMI, IA, IIT, and CLVG). The basic context of the project is establishing the foundations of a new interdisciplinary area called Astroinformatics which risen from the need of ICT methods for preservation and exploitation of the scientific, cultural and historic heritage of astronomical wide-field-plate observations. The main project objective is the usage of ICT methods and instruments for processing, representation and storage of astro-images in specialized database for efficient data retrieval. A web-based system should be developed in view of its future inclusion into the Euro-VO.
ISSSE: (2006-2009): Intelligent sensor systems for security enhancement. Research project VU-MI-204/2006, funded by the National Science Fund of Bulgaria, coordinated by FMI of the Sofia-University. IIT-BAS partnership The project aims the approbation of the new methods and means for the processing of sensor information, and the establishment of a simplified prototype of a system, targeting the problems of security and the war with terrorism. Some of the specific projects tasks are: sensor calibration, 3D reconstruction, object recognition, support of a database of the objects of interest, and other.
Fast Access Methods by Content to Multimedia Databases. IIT-BAS research project I-1306/ 2003, funded by the National Science Fund of Bulgaria, (2003-2006) The project objective was the improvement of the indexing techniques for effective CBIR (Content Based Image Retrieval), so that they are both fast and noise resistant. The project was targeted to meet the needs of the Bulgarian Patent Office, with their large image database of trade marks and service marks.
STEMB: Software System for Replicas Verification of the State Emblem of Bulgaria, two IIT-BAS R&D Projects, funded by the National Committee of Standards and Metrology (1998–1999), (2000-2001) The purpose of the system STEMB is to verify replicas of the State Emblem, i.e. to compare between the State Emblem’s standard and its replica, in order to detect possible inaccuracies and divergences, to estimate them objectively both qualitatively and quantitatively and finally to document them.
Written and Printed Text Recognition – Methods and Information Technology for Cyrillic. IIT-BAS research project I-524/1995, funded by National Science Fund of Bulgarian Ministry of Education and Science, (1996-1998) The system development problems for an effective recognition of text images, representing handwritten and/or printed text in Cyrillic, and more especially in Bulgarian are considered. An argumentation has been proposed for information technology predominantly oriented towards possible transfer of the methods, algorithms and tools, which are already approved in the case of Latin recognition.
Biometric Parameters Based Identification, IIT-BAS research project I-1302/2003, funded by National Science Fund of Bulgarian Ministry of Education and Science, (2003–2005) The goal of the project is to investigate the possibility for reliable authentication of individuals using biometric parameters. Attention is paid to the following modalities: face, signature, voice and moving hand.
Neural Networks for Classification: Theoretical Researches and Application Aspects. IIT-BAS research project I-611/1996 funded by National Science Fund of Bulgarian Ministry of Education and Science, (1996-2000) The project develops new methods for description, learning and minimization of neural networks (NN) topology. The study of NN is based on modern and fast algorithms for classification, optimization and identification. The results are implemented in pattern recognition and dynamic system identification and simulation.
Computer System for CD's Identification, IIT-BAS R&D project funded by the Audio-Accoustic Federation (1998-1999) The aim of the project was to identity the machines producing illegal compact disks. For this a microscope and image processing techniques are used. The system is implemented in the Institute for Forensic Science and Criminology in Sofia.
CAPI: Computer-Aided Portrait Investigation, IIT-BAS R&D project, funded by the Institute of Forensic Science and Criminology in Sofia, (1996-1997) The goal is to assist the expert to identity photographs on the basis of anthropometric points, thus increasing the objectivity of the conclusion on the one hand, and diminishing the expertise expenses. Implemented in the Institute for Forensic Science and Criminology in Sofia.

Selected Publications:
BOOKS
  1. Atanassov K., G. Gluhchev, S. Hadjitodorov, A. Shanon, V. Vassilev, Generalized Nets in Pattern recognition, KvB, Monograph #6, Visual Concepts Pty, Ltd., Australia, 2003.
  2. Atanassov K., G. Gluhchev, S. Hadjitodorov, J. Kasprzyk, A. Shannon, E. Schmidt, V. Vassilev. Generalized Nets, Decision Making and Pattern Recognition, Warsaw School of Inf. Technology, Warsaw, 2006.
PAPERS
Handwriting Analysis and Recognition
  1. Nestorov D., St. Benchev, St. Bonchev, G. Gluhchev. “ABC – Alphabetical Computer-Based Handwriting Investigation System”, Cybernetics and Information Technologies, Vol. 8, #1, Sofia, 2008, pp. 82-88.
  2. Gluhchev G., O. Boumbarov. “Automatic Evaluation of Stroke Slope”, Proc. 1st European Workshop on Biometrics and Identity Management, Roskilde University, May 7-9, , Denmark, 2008, pp. 204-207.
  3. Laskov L., and D. Dimov, "Segmentation of Ancient Neumatic Musical Notation", In: Proceedings of the Int.Conf. A&I’08, 01-04.10.08, Sofia, 2008, pp II.21-24.
  4. Laskov L., and D. Dimov: Color Image Segmentation for Neume Note Recognition, In: Proceedings of the Int.Conf. A&I’07, 03-06.10.07, Sofia, 2007, pp. III.37-41.
Image & Video Analysis and Recognition
  1. Shapiro V., G. Gluhchev, D. Dimov. “Towards a Multinational Car License Plate Recognition System”, Machine Vision and Application Journal, Vol. 17, No. 3, Springer Berlin/Heidelberg, 2006, pp. 173-183.
  2. Gluhchev G., M. Savov, O. Boumbarov. “Classification of classifiers”, iCEST 2007, 24-27 June, Ohrid, 2007, pp. 521-522.
  3. Shapiro V., G. Gluhchev. “Multinational license plate recognition system: Segmentation and Classification”, ICPR’04, Vol. 4/4, Cambridge, 2004, pp. 352-355.
  4. Gluhchev G., S. Bonchev. “A real-time image restoration”, Proc. of CompSysTech’2003, Sofia, pp. III.3-1 – III.3-6.
  5. S. Bonchev, G. Gluhchev. “A face identification method in case of different close-up”, “Recpad2000”, Porto, Portugal, 2000, pp. 103-108.
  6. Dimov D., and A. Marinov: Geometric-Morphological Method for Artifact Noise Isolation in the Image Periphery by a Contour Evolution Tree, In: Proceedings of CompSysTech’2006, June 15-16, V.Tyrnovo, 2006, pp.IIIb.15.1-8, (http://ecet.ecs.ru.acad.bg/cst06/Docs/cp/SIII/IIIB.15.pdf)
  7. Dimov D., and N. Zlateva: Harmonic methods for solid noise filtering from the image periphery, In: Proceedings of CompSysTech’06, June 15-16, 2006, V.Tyrnovo, Bulgaria, pp.IIIb.16.1-8, (http://ecet.ecs.ru.acad.bg/cst06/Docs/cp/SIII/IIIB.16.pdf)
  8. Dimov D., G. Gluhchev, N. Nikolov, I. Burov, E. Kalcheva, S. Bonchev, S. Milanov, and B. Kiossev: The Computer System STEMB for Verification of Replicas of the Bulgarian State Emblem, Cybernetics and Information Technologies, Bulgarian Academy of Sciences, Vol.1, 1, 2001, pp.95-106. (http://www.cit.iit.bas.bg/CIT_01_en/1-01/D_Dimov.html)
  9. Dimov D. T., and G. Y. Gluhchev: A Locally Adaptive Binary-Tree Method for Binarization of Text Images. In: A. C. Downton and S. Impedovo (Eds.), Progress in Handwriting Recognition, World Scientific, London, 1997, pp.575-580.
  10. Dimov D. T.: Natural Language Names Recognition Method - Fault Tolerant to the Most Common Mistypings. In: A. M. Ramsay (ed.), Artificial Intelligence: Methodology, Systems, Applications (AIMSA'96), IOS Press, 1996, pp.320-329.
Speech Analysis and Speaker Recognition
  1. Ouzounov A., A Robust Feature for Speech Detection, Cybernetics and Information Technologies, vol.4, No.2, 2004, pp.3-14.
  2. Ouzounov A., BG-SRDat: A Corpus in Bulgarian Language for Speaker Recognition over Telephone Channels, Cybernetics and Information Technologies, vol.3, No.2, 2003, pp.101-108.
  3. Ouzounov A. Text-Independent Speaker Identification using a Hybrid Neural Network and Conformity Approach, In Proceedings of The IEEE International Conference on Neural Networks, 1997, Houston, vol.IV, pp.2098-2102.
  4. Ouzounov A., Clipped LPC Cepstrum and Its Application to Text-Independent Speaker Identification, Cybernetics and Information Technologies, vol.1, No.2, 2001, pp.19-29.
  5. Ouzounov A., Robust Features for Speech Detection – A Comparative Study, In Proceedings of The International Conference on Computer Systems and Technologies, CompSysTech’2005, pp.IIIA.19-1 – IIIA.19-6.
  6. Dimov, D. and I. Azmanov: Heuristic improvements of HMM use in isolated word speech recognition, Cybernetics and Information Technologies, Bulgarian Academy of Sciences, Vol.7, 3, 2007, pp.73-88. (http://www.cit.iit.bas.bg/CIT_07/v7-3/73-88.pdf)
Content Based Image/Object Retrieval (CBIR/CBOR)
  1. Dimov D.T., N. Zlateva, and A. Marinov: CBIR Approach to Face Recognition, In Proc. of A&I’08 Conf., 01-04.10.2008, Sofia, Workshop on Multisensor signal, image and data processing, 02.10.2008, pp. IV.21-26.
  2. Dimov D.: A Polar-Fourier-Wavelet Transform for Effective CBIR, In: Proceedings of the ADMKD'07, 02.10.07, Varna (in the frames of 11-th Int. Conf. ADBIS'07, 29.09-03.10.07, Varna), 2007, pp.107-118.
  3. Dimov D.: Rapid and Reliable Content Based Image Retrieval, In: Proceedings of NATO ASI on Multisensor Data and Information Processing for Rapid and Robust Situation and Threat Assessment, May 16-27, 2005, Albena-Bulgaria, IOS Press, 2007, pp.384-395.
  4. Dimov D., A. Marinov, and N. Zlateva: CBIR Approach to the Recognition of a Sign Language Alphabet, Conf. Proceedings of CompSysTech’2007, June 14-15, Rousse, 2007, pp.V.2.1-9, (http://ecet.ecs.ru.acad.bg/cst07/Docs/cp/sV/V.2.pdf)
  5. Dimov D.: Harmonic and Wavelet Transform Combination for Fast Content Based Image Retrieval, Int. Conference on PDE Methods in Applied Mathematics and Image Processing, Sep. 7-10, 2004, Sunny Beach, Bulgaria.(http://www.math.bas.bg/~kounchev/2004_conference/HomePage2004.html)
  6. CBIR/CBOR application(s) based on a FANTIR technology, Brokerage Event on Information Technologies during 20th INFOSYSTEM, 30.09-01.10.2006, Thessaloniki, Greece, http://www.innovationrelay.net/bemt/catalog.cfm?status=2&whattodisp=profiledetails&eventid=1405&pr_id=14255&CFID=24699&CFTOKEN=75182066
Biometric based identification
  1. Atanassov K., Gluhchev G., S. Hadjitodorov, M. Savov, E. Szmidt. “Signature based person verification – a generalized net model”, Current Issues in Data and Knowledge Engineering (B.D.Baets, et all, Eds), Academicka Ofycina Widawnicza EXIT, Warszawa, 2004, pp. 33-41.
  2. Savov M., G. Gluhchev. “Signature verification via Hand-Pen motion investigation”, Proc. Int. Conf. “Recent Advances in Soft Computing”, Canterbury, 2006, pp. 490-495.
  3. Shanon A., G. Gluhchev, K. Atanassov, S. hadjitodorov, V. Vassilev. “A generalized net model of biometrics-based access permission”, Comptes rendus, Vol. 60, No 11, 2007, pp.1157-1162.
  4. Atanassov K., G. Gluhchev, S. Hadjitodorov, A. Shanon, V. Vassilev. “Application of generalized nets in biometrics”, Comptes Rendus de l”Academie Bulgare des Sciences, vol. 56, No 5, Sofia, 2003, pp. 13-18.
  5. Gluhchev G., M. Savov, O. Boumbarov, D. Vassileva. A New Approach to Signature Based Authentication”, 2nd Int. Conf. on Biometrics, Seoul, 26-29 August, 2007, pp. 594-603.
  6. Ñàâîâ Ì., Èäåíòèôèêàöèÿ íà ëèöà ïðè ïîëàãàíå íà ïîäïèñ, Äèñåðòàöèÿ çà ÍÎÑ “äîêòîð”, ÈÈÒ-ÁÀÍ, 2007.
Neural Network Based Classification
  1. Ilchev, V., (2002). A Model of a Neural Cell's Group in the Multidimensional Hough Space. Comptes rendus de l'Academie Bulgare des Sciences, Tome 55, ¹ 1, 57- 62.
  2. Ilchev, V., S. Koynov., (2000). Properties of the Model of a Neural Cells' Group with Multiple Synaptic Connections. Comptes rendus de l'Academie Bulgare des Sciences, Tome 53, V.8, 41- 44.
  3. Ilchev, V., (2003). Transformation of Lineary Separable Sets in Hough Spase. Comptes rendus de l'Academie Bulgare des Sciences, Tome 56, ¹ 1, 31- 36.
  4. Ilchev, V., S. Koynov, Z. Ilcheva, I. Baruch., (1997). A Neurophysiological Model of the Retina with Neural Networks. Proceedings of the 4th Seminar on Neural Network Applications in Electrical Engineering "NEUREL 97", Belgrade, Yugoslavia, 188- 192.
  5. V. Ilchev, V. Zl. Ilcheva., (2006). A Self- Organizing Neural Network for Variable Convex Clusterization, Cybernetics and Information Technologies, (CIT), No 1, 17-23.
  6. V. Ilchev., Zl. S. Ilcheva., (2008). Basic Properties of the Classification Hyperplanes Represented by Extremum Points in the Multidimensional Hough Space, Cybernetics and Information Technologies (CIT), Vol. 8, No 4, 27-39.
Computer Assisted Medical Diagnostics
  1. G. Gluhchev, S. Shalev, "The systematic error detection as a classification problem", Pattern Recognition Letters, Vol. 17 (12), 1996, pp. 1233-1238.
  2. G. Gluhchev. “The magnitude of treatment field set-up parameter correction in radiation therapy”, Radiother. Oncol. Vol. 48,1998, pp.79-82.
  3. E. Kalcheva, G.Gluhchev. “Investigation of handwritten graphics from patients with neurological diseases”, Proc. Int. Conf. On Computer Systems and Technologies – CompSysTech’2001, Sofia, 2001, pp.III.31-1 – III.31-5.
Other
  1. Dimov D., N. Dokev, P. Asenova, and S. Ivanov: On Generalized Evaluation of PhD Competences in Computing, In: Proceedings of 3-d e-Learning Conf. Sept 7-8, 2006, Coimbra, Portugal, ISBN 978-989-20-0350-4, pp. 2.3.1-8 (http://elconf06.dei.uc.pt/pdfs/paper33.pdf)

Last update: April 2009