«INFORMATICS AND APPLICATIONS»
Scientific journal
Volume 4, Issue 1, 2010

Content | Bibliography | About  Authors

Abstract and Keywords.

ON-LINE INFORMATION MODEL BUILDING OF THE EARTH POLE MOTION BY LINEAR AND LINEARIZED FILTERS.

• I.N. Sinitsyn   IPI RAN, sinitsin@dol.ru
• V. I. Sinitsyn   IPI RAN, vsinitsin@ipiran.ru
• E.R. Korepanov   IPI RAN, ekorepanov@ipiran.ru
• V.V. Belousov   IPI RAN, vbelousov@ipiran.ru
• N.N. Semendyaev   IPI RAN, nsemendyaev@ipiran.ru

Abstract:  Practical problems of on-line continuous and discrete filtering (estimation of state and parameters) of the Earth pole motion by linear and linearized filters are considered. Instrumental algorithms and software tools in MATLAB are discussed. Test examples and computer experiments based on astronometric measurements are presented.

Keywords:  astronometric measurements; instrumental algorithms; software tools; information model; linear and linearized filters; generalized Kalman filters; on-line filtering; Kalman-Bucy filters; Pugachev filters; fluctuations of the Earth pole

ASYMPTOTIC DISTRIBUTIONS OF BASIC STATISTICS IN GEOMETRIC REPRESENTATION FOR HIGH-DIMENSIONAL DATA AND THEIR ERROR BOUNDS.

• Y. Kawaguchi   Department ofMathematics, Graduate School of Science and Engineering, Chuo University, Bunkyo-ku, Tokyo, Japan, n15007@gug.math.chuo-u.ac.jp
• V.V. Ulyanov   Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, vulyan@gmail.com
• Y. Fujikoshi   Department of Mathematics, Graduate School of Science and Engineering, Chuo University, Bunkyo-ku, Tokyo, Japan, fuji@math.sci.hiroshima-u.ac.jp

Abstract: In geometric representation of n observations on p variables, it is necessary to examine asymptotic behaviors of the three statistics; the length of a p-dimensional observation vector, the distance between two independent observation vectors, and the angle between these observation vectors. Hall et al. (2005) found the asymptotic values of these three statistics in high-dimensional framework when the dimension p tends to infinity, while the sample size n is fixed. In this paper, their results are extended by deriving asymptotic expansions of the distributions of the three statistics. Further, computable error bounds for the limiting distributions of the length and the distance were obtained. These results will be useful to obtain statistical insights in middle - as well as in high-dimensional data sets.

Keywords:  asymptotic expansions; error bounds; high-dimensional data; geometric representation

ASYMPTOTIC EXPANSION FOR THE POWER OF TEST BASED ON SAMPLE MEDIAN IN THE CASE OF LAPLACE DISTRIBUTION.

• V. E. Bening   Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, bening@yandex.ru
• A. V. Sipina   Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, anna@sipin.ru

Abstract:  Using asymptotic expansions, the formula for the limit of the difference between the powers of the test based on the sample median and the most powerful test for the case of Laplace distribution was obtained.

Keywords:  sample median; asymptotic expansion; power function; Laplace distribution

ORGANIZATION OF USERS' MANAGEABLE ACCESS TO HETEROGENEOUS DEPARTMENTAL INFORMATIONAL RESOURCES.

• G. Y. Ilyushin   IPI RAN, ilushin@ipiran.ru
• I.A. Sokolov   IPI RAN, isokolov@ipiran.ru

Abstract:  The paper is devoted to the realization issues of application interoperability and user access to data repositories management during IT-infrastructure renovation in large enterprises. For the governed evolution of the legacy management information systems and data repositories without interruption of existing systems operation, a technology is proposed for creating the middleware infrastructure based on hybrid software/hardware solutions.
The middleware based on the service-oriented architecture and Web Services technologies solves the problem of interoperability of heterogeneous applications a swell as the problem of providing a centralized model of management of user access to heterogeneous data repositories on the basis of formalized roles. Besides interoperability issues, the middleware infrastructure enables to solve a broad spectrum of problems related to a provision of information security of the required level.

Keywords:  interoperability; access management; middleware; Web Services; legacy systems; metadata

PETROLEUM AS A CARRIER OF INFORMATION ON ITS ORIGIN, STRUCTURE, AND EVOLUTION.

• R.B. Seiful-Mulukov   IPI RAN, rust@ipiran.ru

Abstract:  One of the most complicated and still unambiguously unresolved problems concerning the origin of petroleum parent substance is considered. It is proved that the information on petroleum is contained in the petroleum itself and the petroleum as a complex system can be treated within the frame of the laws of informatics.

Keywords:  petroleum; petroleum geneses; petroleum information carrier; petroleum information content; information content evaluation

MODELING OF ELASTIC DEFORMATIONS IMPACT ON FINGERPRINT RECOGNITION PERFORMANCE.

• A.R. Arutyunyan   Nuclear Safety Institute of the Russian Academy of Sciences, artem@ibrae.ac.ru

Abstract:  Nuclear Safety Institute of the Russian Academy of Sciences, artem@ibrae.ac.ru The problem of registration and modeling of distortions in biometric systems is considered. A biometric system performance is measured by false acceptance and false rejection rate which can be calculated from score distributions in genuine and impostor matches. The statistical moments are used to describe these score distributions with few parameters. Distortions shift the distributions and, consequently, moments. Based on it, an arbitrary distortion is modeled as shifts in score distribution. Finally, the experiments with modeling of impact of elastic deformation on fingerprint recognition were carried out.

Keywords:  biometrics; operational testing; nonlinear fingerprint deformations

MATHEMATICAL MODELS OF FINGERPRINT IMAGE ON THE BASIS OF LINES DESCRIPTION.

• V.Yu. Gudkov   Chelyabinsk State University, Department of Applied Mathematics, diana@sonda.ru

Abstract:  The mathematical models of fingerprint based on the lines as topological vectors and ridge count are presented. They are stored in the template with the list of minutiae. The templates are used to identify the fingerprint.

Keywords:  fingerprint; minutiae; topology; event detector; ridge count

AUTOMATED QUALITY TESTING OF DIGITAL IMAGERY FOR PERSONAL DOCUMENTS.

• S. L. Karateev   FGUP "GosNIIAS," goga@gosniias.ru
• I. V. Beketova   FGUP "GosNIIAS," irus@gosniias.ru
• M. V. Ososkov   FGUP "GosNIIAS"
• V.A. Knyaz   FGUP "GosNIIAS"
• Yu. V. Vizilter   FGUP "GosNIIAS"
• A. V. Bondarenko   FGUP "GosNIIAS"
• S. Yu. Zheltov   FGUP "GosNIIAS"

Abstract:  The software-hardware system for digital face imagery acquisition and testing for requirements of ISO/IEC FCD 19794-5 standard is described. System contains the following algorithmic modules: face detector; color and intensity characteristics estimator; opened/closed eyes detector; glasses detector, reflexes, shines, and shadows detector; face features detector (nose, brows, mouth); face slope/rotation detector. The precision of face orientation estimation based on monocular digital imagery is addressed. The approach for precision estimation is developed based on comparison of synthesized facial two-dimensional images and scanned face three-dimensional model.

Keywords:  biometrics; personal documents; face imagery; face detection; boosting; three-dimensional reconstruction; three-dimensional modeling

AN ALGORITHMFOR CLOTHES-BASED HUMAN RECOGNITION IN VIDEO.

• V. S. Konushin   M.V. Keldysh Institute of Applied Mathematics of the Russian Academy of Sciences; M.V. Lomonosov Moscow State University, vadim@graphics.cs.msu.ru
• G.R. Krivovyaz   M.V. Lomonosov Moscow State University, gkrivovyaz@graphics.cs.msu.ru
• A. S. Konushin   M.V. Lomonosov Moscow State University, ktosh@graphics.cs.msu.ru

Abstract:  The new algorithm for clothes-based human recognition in video is presented. The algorithm is based on random patches classification with a random forest. The main advantage of the algorithm is the fact that it does not rely on human mask; therefore, it is able to work with video with arbitrary complex background. Experimental results are reported for a test set, which was acquired using authors' video surveillance system.

Keywords:  video-based human recognition; machine learning; random forest; background subtraction

HERMITE TRANSFORM BASED IRIS KEY POINTS SELECTION AND ANALYSIS.

• E. A. Pavelyeva   Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, paveljeva@yandex.ru
• A. S. Krylov   Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, kryl@cs.msu.ru

Abstract:  Key points selection method based on local Hermite transform for iris recognition has been designed. Use of iris key points enables to obtain good recognition results with small storage capacity.

Keywords:  biometrics; iris; Hermite transform; key points

DRAWING UP OF AN IDENTIKIT USING EVOLUTIONARY MORPHING AND QUALIMETRY METHOD.

• V. Protasov   Russian Center of Computing for Physics and Technology, Protvino, protonus@yandex.ru

Abstract: The new information technology for drawing up of an identikit is resulted. It is based on genetic algorithms and work of group of witnesses. The model of "the virtual witness" has been developed. The model is intended for calculation of accuracy of drawing up of an identikit. It includes ability of the person to drawing persons and the ability to compare different individuals according to their similarity to the original. The fatigue of a witness after a long work on the recognition and comparison of individuals is counted. The model provides setting method and qualimetry of evolutionary morphing.

Keywords:  identikit; evolutionary morphing; genetic algorithms; decision making; qualimetry