Geodesic
A simple upper bound to the Bayes error probability for feature selection
Kybernetika, Tome 34 (1998) no. 4, pp. 387-392
Cet article a éte moissonné depuis la source Czech Digital Mathematics Library

Voir la notice de l'article

In this paper, feature selection in multiclass cases for classification of remote-sensing images is addressed. A criterion based on a simple upper bound to the error probability of the Bayes classifier for the minimum error is proposed. This criterion has the advantage of selecting features having a link with the error probability with a low computational load. Experiments have been carried out in order to compare the performances provided by the proposed criterion with the ones of some of the widely used feature-selection criteria presented in the remote-sensing literature. These experiments confirm the effectiveness of the proposed criterion, which performs slightly better than all the others considered in the paper.
In this paper, feature selection in multiclass cases for classification of remote-sensing images is addressed. A criterion based on a simple upper bound to the error probability of the Bayes classifier for the minimum error is proposed. This criterion has the advantage of selecting features having a link with the error probability with a low computational load. Experiments have been carried out in order to compare the performances provided by the proposed criterion with the ones of some of the widely used feature-selection criteria presented in the remote-sensing literature. These experiments confirm the effectiveness of the proposed criterion, which performs slightly better than all the others considered in the paper.
Classification : 62H30, 68G10, 68T10
Keywords: Bayes error probability; feature selection 
@article{KYB_1998_34_4_a5,
     author = {Bruzzone, Lorenzo and Serpico, Sebastiano B.},
     title = {A simple upper bound to the {Bayes} error probability for feature selection},
     journal = {Kybernetika},
     pages = {387--392},
     year = {1998},
     volume = {34},
     number = {4},
     zbl = {1274.68377},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a5/}
}
TY  - JOUR
AU  - Bruzzone, Lorenzo
AU  - Serpico, Sebastiano B.
TI  - A simple upper bound to the Bayes error probability for feature selection
JO  - Kybernetika
PY  - 1998
SP  - 387
EP  - 392
VL  - 34
IS  - 4
UR  - http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a5/
LA  - en
ID  - KYB_1998_34_4_a5
ER  - 
%0 Journal Article
%A Bruzzone, Lorenzo
%A Serpico, Sebastiano B.
%T A simple upper bound to the Bayes error probability for feature selection
%J Kybernetika
%D 1998
%P 387-392
%V 34
%N 4
%U http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a5/
%G en
%F KYB_1998_34_4_a5
Bruzzone, Lorenzo; Serpico, Sebastiano B. A simple upper bound to the Bayes error probability for feature selection. Kybernetika, Tome 34 (1998) no. 4, pp. 387-392. http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a5/

[1] Bruzzone L., Roli F., Serpico S. B.: An extension of the Jeffreys–Matusita distance to multiclass cases for feature selection. IEEE Trans. Geoscience Remote Sensing 33 (1995), 6, 1318–1321 | DOI

[2] Bruzzone L., Serpico S. B.: Feature selection in multiclass cases: a proposal and an experimental investigation. In: Proceedings of the 1st Workshop on Statistical Techniques in Pattern Recognition, Prague 1997, pp. 19–24

[3] Fukunaga K.: Introduction to Statistical Pattern Recognition. Second edition. Academic, New York 1990 | MR | Zbl

[4] Liu S. S., Jernigan M. E.: Texture analysis and discrimination in additive noise. Computer Vision Image Processing 49 (1990), 52–67 | DOI

[5] Swain P. H., Davis S. M.: Remote Sensing: The Quantitative Approach. McGraw–Hill, New York 1994

[6] Tou J. T., Gonzalez R. C.: Pattern Recognition Principles. Addison–Wesley, London 1974 | MR | Zbl