Geodesic
Using the beta distribution for the analysis of informative value of features and for improving the efficiency of decision rule for texture images recognition
Vestnik Udmurtskogo universiteta. Matematika, mehanika, kompʹûternye nauki, no. 3 (2014), pp. 134-141 Cet article a éte moissonné depuis la source Math-Net.Ru

Voir la notice de l'article

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, informative value of features, binary classifier.
Mots-clés : beta distribution 
@article{VUU_2014_3_a11,
     author = {V. N. Milich and V. A. Smetanin},
     title = {Using the beta distribution for the analysis of informative value of features and for improving the efficiency of decision rule for texture images recognition},
     journal = {Vestnik Udmurtskogo universiteta. Matematika, mehanika, kompʹ\^uternye nauki},
     pages = {134--141},
     year = {2014},
     number = {3},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VUU_2014_3_a11/}
}
TY  - JOUR
AU  - V. N. Milich
AU  - V. A. Smetanin
TI  - Using the beta distribution for the analysis of informative value of features and for improving the efficiency of decision rule for texture images recognition
JO  - Vestnik Udmurtskogo universiteta. Matematika, mehanika, kompʹûternye nauki
PY  - 2014
SP  - 134
EP  - 141
IS  - 3
UR  - http://geodesic.mathdoc.fr/item/VUU_2014_3_a11/
LA  - ru
ID  - VUU_2014_3_a11
ER  - 
%0 Journal Article
%A V. N. Milich
%A V. A. Smetanin
%T Using the beta distribution for the analysis of informative value of features and for improving the efficiency of decision rule for texture images recognition
%J Vestnik Udmurtskogo universiteta. Matematika, mehanika, kompʹûternye nauki
%D 2014
%P 134-141
%N 3
%U http://geodesic.mathdoc.fr/item/VUU_2014_3_a11/
%G ru
%F VUU_2014_3_a11
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. Vestnik Udmurtskogo universiteta. Matematika, mehanika, kompʹûternye nauki, no. 3 (2014), pp. 134-141. http://geodesic.mathdoc.fr/item/VUU_2014_3_a11/

[1] Lyal'ko V. I., Popov M. A., Podorvan V. N., Sakhatskii A. I., “Planar objects on the multispectral space imagery classification technique on the basis of the consequent information merging”, Current problems in remote sensing of the Earth from space, 1:2 (2004), 88–95 (in Russian)

[2] Smetanin V. A., Milich V. N., “Investigation of bispectral co-occurrence matrices for multispectral data analysis”, Vestn. Izhevsk. Gos. Tekhn. Univ., 2012, no. 4, 123–126 (in Russian)

[3] Plotnikov D. E., Bartalev S. A., Zharko V. O., Mihailov V. V., Prosyannikova O. I., “An experimental assessment of crop types recognisability using time-series of intra-seasonal spectral reflectance measurements by satellite sensor”, Current problems in remote sensing of the Earth from space, 8:1 (2011), 199–208 (in Russian)

[4] Tou J. T., Gonzalez R. C., Pattern recognition principles, AWPC Inc., 1974 | MR | MR

[5] Pavlov I. N., Ushanov S. V., “Investigation of tree diameter distributions by methods of distribution mixes”, Vestn. Sib. Gos. Tekhnol. Univ., 2005, no. 1, 38–46 (in Russian)

[6] Elderton W. P., Frequency curves and correlation, London, 1906 https://ia700401.us.archive.org/2/items/frequencycurvesc00elderich/frequencycurvesc00elderich.pdf

[7] Nikulin M. S., “Chi-square test for continuous distributions with shift and scale parameters”, Theory of Probability and Its Applications, 18:3 (1974), 559–568 | DOI | MR | Zbl