Loading [MathJax]/jax/output/SVG/config.js
Vestnik Udmurtskogo Universiteta. Matematika. Mekhanika. Komp'yuternye Nauki
RUS  ENG    JOURNALS   PEOPLE   ORGANISATIONS   CONFERENCES   SEMINARS   VIDEO LIBRARY   PACKAGE AMSBIB  
General information
Latest issue
Archive
Impact factor

Search papers
Search references

RSS
Latest issue
Current issues
Archive issues
What is RSS



Vestn. Udmurtsk. Univ. Mat. Mekh. Komp. Nauki:
Year:
Volume:
Issue:
Page:
Find






Personal entry:
Login:
Password:
Save password
Enter
Forgotten password?
Register


Vestnik Udmurtskogo Universiteta. Matematika. Mekhanika. Komp'yuternye Nauki, 2014, Issue 3, Pages 134–141 (Mi vuu446)  

This article is cited in 1 scientific paper (total in 1 paper)

COMPUTER SCIENCE

Using the beta distribution for the analysis of informative value of features and for improving the efficiency of decision rule for texture images recognition

V. N. Milich, V. A. Smetanin

Physical Technical Institute, Ural Branch of the Russian Academy of Sciences, ul. Kirova, 132, Izhevsk, 426000, Russia
Full-text PDF (236 kB) Citations (1)
References:
Abstract: The problems of texture images classification and feature space reduction are considered. The multialternative classification is reduced to a binary one-dimensional problem, for which it is possible to use Bayesian approach with one-dimensional estimates of distribution. Hypothesis of beta-distribution for one separate class is introduced. Distribution parameters are estimated by method of moments. To estimate four parameters, analytic expressions and statistic estimations of first four moments of this distribution are needed. After that the hypothesis of distribution is verified by Pearson's criterion. It is established experimentally that beta-distribution model is generally applicable to estimate feature values distribution. It is concluded that such checks are needed for every learn sample. Feature effectiveness is estimates by analysis of degree of estimated distributions intersection. The cross-correlation of selected features is considered. A method for estimating the informative value of feature is introduced. It is based on the minimum average error probability for one feature and on mutual uncorrelatedness for a system of features. For each pair of classes we build feature system using the algorithm for estimation of informative value. The classifier, which uses feature systems and makes decision on the basis of density estimation by beta-distribution model on the stage of the binary problem, is built. This classifier also aggregates the final decision from the results of binary problems decisions.
Keywords: texture images, beta distribution, informative value of features, binary classifier.
Received: 19.06.2014
Document Type: Article
UDC: 519.233+528.854
MSC: 62F03, 68T45
Language: Russian
Citation: V. N. Milich, V. A. Smetanin, “Using the beta distribution for the analysis of informative value of features and for improving the efficiency of decision rule for texture images recognition”, Vestn. Udmurtsk. Univ. Mat. Mekh. Komp. Nauki, 2014, no. 3, 134–141
Citation in format AMSBIB
\Bibitem{MilSme14}
\by V.~N.~Milich, V.~A.~Smetanin
\paper Using the beta distribution for the analysis of informative value of features and for improving the efficiency of decision rule for texture images recognition
\jour Vestn. Udmurtsk. Univ. Mat. Mekh. Komp. Nauki
\yr 2014
\issue 3
\pages 134--141
\mathnet{http://mi.mathnet.ru/vuu446}
Linking options:
  • https://www.mathnet.ru/eng/vuu446
  • https://www.mathnet.ru/eng/vuu/y2014/i3/p134
  • This publication is cited in the following 1 articles:
    1. A. V. Kuchuganov, D. R. Kasimov, V. N. Kuchuganov, “Modelirovanie rassuzhdenii pri poiske ob'ektov na izobrazheniyakh”, Vestn. Udmurtsk. un-ta. Matem. Mekh. Kompyut. nauki, 30:3 (2020), 497–512  mathnet  crossref
    Citing articles in Google Scholar: Russian citations, English citations
    Related articles in Google Scholar: Russian articles, English articles
    Вестник Удмуртского университета. Математика. Механика. Компьютерные науки
     
      Contact us:
    math-net2025_01@mi-ras.ru
     Terms of Use  Registration to the website  Logotypes © Steklov Mathematical Institute RAS, 2025