Authors Yu. A. Bryuhomitsky
Month, Year 05, 2018 @en
Index UDC 004.067
Abstract An immunological approach is proposed to solve the problem of recognition of dynamic biometric signals, based on the principles of massively parallel decentralized data processing used in artificial immune systems. A special feature of approach is presentation of dynamic biometrics signals by sequences of informational units of a certain format, with their following RTP processing on the basis of the immunological clonal selection model with positive selection. As an informational units are syntactically related fragments of the text of the corresponding modality. They are represented by multidimensional vectors in the workspace of features. In the training phase, an initial population of detectors is created in the metric of the vectors of the investigated sequence of biometric data. Then, according to the principle of positive selection, the detectors of the initial population are identified, which in the feature space are closest to the areas of distribution of the corresponding biometric data. The proximity of the vectors in the feature space models the property of the affinity of the immune system cells. Revealed detectors, using the iterative procedure, are subjected to cloning, hypermutation and selection and ultimately form a population of immune memory detectors. The training procedure is stopped when the specified maximum size of the population of the immune memory detectors is reached.In the recognition phase, the elements of the analyzed biometric data sequence are compared with the memory population detectors using the Euclid proximity measure. The critical level of proximity defines the boundary for the decision "well-known/stranger" making by the system and is specified based on the permissible errors of the first kind. To identify "well-known/stranger", a statistical approach is used, in which the frequency of the critical proximity condition is controlled, which determines the statistical probability of belonging to the analyzed biometry to "stranger". The proposed approach within the framework of the immunological presentation allows to generalize essentially different methods of personality identification by the dynamic biometric parameters of different modalities. The positive differences of the proposed approach are: the possibility of text-independent analysis of dynamic biometrics of any modality, arbitrary volume and content; continuous assessment of the biometric data in RTP with the possibility of timely decision-making on the presence of "stranger"; the use of an immunological model that fits well with most of the tasks of dynamic personality identification, which allows to reduce significantly the number of detectors required for effective identification of a person.

Download PDF

Keywords Text-independent personality identification; dynamic biometry; artificial immune systems; vector data representation; immune model of clonal selection.
References 1. Akhmad Kh.M., Zhirkov V.F. Vvedenie v tsifrovuyu obrabotku rechevykh signalov [ntroduction to digital speech signal processing]. Vladimir: Izd-vo Vladim. gos. un-ta, 2007, 192 p.
2. Matveev Yu.N. Tekhnologii biometricheskoy identifikatsii lichnosti po golosu i drugim modal'nostyam [Technologies of biometric identification by voice and other modalities], Vestnik MGTU im. N.E. Baumana. Ceriya «Priborostroenie» [Herald of the Bauman Moscow State Technical University. Series Instrument Engineering], 2012, No. 2, pp. 46-61.
3. Campbell W., Assaleh K., Broun C. Speaker recognition with polynomial classifiers, IEEE Trans. Speech Audio Process, 2002, Vol. 10, No. 4, pp. 205-212.
4. Anisimova E.S. Identifikatsiya onlayn-podpisi s pomoshch'yu okonnogo preobrazovaniya Fur'e i radial'nogo bazisa [Identification of an online signature using a window Fourier transform and a radial basis], Komp'yuternye issledovaniya i modelirovanie [Computer studies and modeling], 2014, Vol. 6, No. 3, pp. 357-364.
5. Jain A.K., Friederike D.G., Connel S.D. On-line signature verification, Pattern Recognition, 2002, Vol. 35 (12), pp. 2963-2972.
6. Plamondon R., Srihari S. On-line and Off-line Handwriting Recognition: A Comprehensive Survey, IEEE Trans. PAMI, 2000, Vol. 22 (1), pp. 63-84.
7. Ivanov A.I. Biometricheskaya identifikatsiya lichnosti po dinamike podsoznatel'nykh dvizheniy: monografiya [Biometric identification of personality by the dynamics of subconscious movements: monograph]. Penza: Izd-vo Penz. gos. un-ta, 2000, 188 p.
8. Bryukhomitskiy Yu.A., Kazarin M.N. Sistema autentifikatsii lichnosti po pocherku [Handwriting authentication system], Sb. trudov nauchno-prakticheskoy konferentsii s mezhdunarodnym uchastiem «Informatsionnaya bezopasnost'» [Proceedings of scientific-practical conference with international participation "Information security"]. Taganrog: Izd-vo TRTU, 2002, pp. 22-29.
9. Maznichenko N.I. Gvozdenko M.V. Analiz vozmozhnostey sistem avtomaticheskoy identifikatsii klaviaturnogo pocherka [Analysis of the possibilities of automatic identification systems of keyboard handwriting], Vestnik Natsional'nogo tekhnicheskogo universiteta «Khar'kovskiy politekhnicheskiy institut». Seriya «Informatika i modelirovanie» [Bulletin of the national technical University "Kharkiv Polytechnic Institute". Series "Informatics and modeling"], 2008, Issue No. 24, pp. 77-82.
10. Skubitskiy A.V. Analiz primenimosti metoda rekonstruktsii dinamicheskikh sistem v sistemakh biometricheskoy identifikatsii po klaviaturnomu pocherku [Analysis of the applicability of the method of reconstruction of dynamic systems in systems of biometric identification by keyboard handwriting], Infokommunikatsionnye tekhnologii [Information and communication technology], 2008, Vol. 6, No. 1, pp. 51-53.
11. Skubitskiy A.V. Analiz primenimosti metoda rekonstruktsii dinamicheskikh sistem v sistemakh biometricheskoy identifikatsii po klaviaturnomu pocherku [The method of biometric identification of the user at the keyboard handwriting based on the decomposition of the Haar and proximity measure of the Hamming], Infokommunikatsionnye tekhnologii [Izvestiya TSURE], 2008, Vol. 6, No. 1, pp. 51-53.
12. Bryukhomitskiy Yu.A. Tsepochnyy metod klaviaturnogo monitoringa [Chained method keyboard monitoring], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2009, No. 11 (100), pp. 135-145.
13. Dasgupta D. Artificial Immune Systems and Their Applications, Ed., Springer-Verlag, 1999.
14. De Castro L.N., Timmis, J.I. Artificial Immune Systems: A New Computational Intelligence Approach, London: Springer-Verlag, 2000, 357 p.
15. Hofmeyr S. and Forrest S. Architecture for an Artificial Immune System, Evolutionary Computation, 2000, No. 8 (4), pp. 443-473.
16. Forrest S., Perelson A.S., Allen L., Cherukuri R. Self-nonself discrimination in a computer, Research in Security and Privacy: Proceedings of IEEE Computer Society Symposium. Los Alamitos: IEEE Computer Society Press, 1994, pp. 202-212.
17. Ji Z., Dasgupta D. Real-valued negative selection algorithm with variable-sized Detectors, Proceedings of the Genetic and Evolutionary Computation, Seattle. 2004, Springer. Verlag: Seattle, WA, USA, 2004, pp. 287-298.
18. Ji Z., Dasgupta D. Revisiting negative selection algorithm, Evolutionary Computation, 2007, Vol. 15, No. 2 (Summer), pp. 223-251.
19. De Castro L.N., Von Zuben F.J. The Clonal Selection Algorithm with Engineering Applications, submitted to GECCO’00, 2000, pp. 36-37.
20. De Castro L.N., Von Zuben F.J. Learning and optimization using the clonal se-lection principle, IEEE Transactions on Evolutionary Computation, Special Issue on Artificial Immune Systems, 2002, Vol. 6, No. 3, pp. 239-251.
21. Bryukhomitskiy Yu.A., Kazarin M.N. Metody mnogosvyaznogo predstavleniya klaviaturnogo pocherka [Methods of multi-linked representation of keyboard handwriting], Materialy III Mezhdunarodnoy konferentsii «Nelokal'nye kraevye zadachi i rodstvennye problemy matematicheskoy biologii, informatiki i fiziki [Materials of the III International conference " Nonlocal boundary value problems and related problems of mathematical biology, Informatics and physics]. Nal'chik, 5-8 dekabrya 2006, pp. 68-69.
22. Bryukhomitskiy Yu.A. Immunologichekiy metod verifikatsii rukopisi s ispol'zovaniem vektornogo predstavleniya dannykh [Immunological method of manuscript verification using vector data representation], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2016, No. 9 (182), pp. 50-57.
23. Bryukhomitskiy Yu.A. Klaviaturnyy monitoring na osnove immunologicheskogo klonirovaniya [Keyboard monitoring based on immunological cloning], Bezopasnost' informatsionnykh tekhnologiy [Information technology security], 2016, No. 4 (40), pp. 5-11.

Comments are closed.