Geodesic
On selecting the best features in a noisy environment
Kybernetika, Tome 34 (1998) no. 4, pp. 411-416 Cet article a éte moissonné depuis la source Czech Digital Mathematics Library

Voir la notice de l'article

This paper introduces a novel method for selecting a feature subset yielding an optimal trade-off between class separability and feature space dimensionality. We assume the following feature properties: (a) the features are ordered into a sequence, (b) robustness of the features decreases with an increasing order and (c) higher-order features supply more detailed information about the objects. We present a general algorithm how to find under those assumptions the optimal feature subset. Its performance is demonstrated experimentally in the space of moment-based descriptors of 1-D signals, which are invariant to linear filtering.
This paper introduces a novel method for selecting a feature subset yielding an optimal trade-off between class separability and feature space dimensionality. We assume the following feature properties: (a) the features are ordered into a sequence, (b) robustness of the features decreases with an increasing order and (c) higher-order features supply more detailed information about the objects. We present a general algorithm how to find under those assumptions the optimal feature subset. Its performance is demonstrated experimentally in the space of moment-based descriptors of 1-D signals, which are invariant to linear filtering.
Classification : 62H30, 62H99, 62M20, 65C60, 68T10
Keywords: Mahalanobis distance; 1-D signals 
@article{KYB_1998_34_4_a9,
     author = {Flusser, Jan and Suk, Tom\'a\v{s}},
     title = {On selecting the best features in a noisy environment},
     journal = {Kybernetika},
     pages = {411--416},
     year = {1998},
     volume = {34},
     number = {4},
     zbl = {1274.62433},
     language = {en},
     url = {http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a9/}
}
TY  - JOUR
AU  - Flusser, Jan
AU  - Suk, Tomáš
TI  - On selecting the best features in a noisy environment
JO  - Kybernetika
PY  - 1998
SP  - 411
EP  - 416
VL  - 34
IS  - 4
UR  - http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a9/
LA  - en
ID  - KYB_1998_34_4_a9
ER  - 
%0 Journal Article
%A Flusser, Jan
%A Suk, Tomáš
%T On selecting the best features in a noisy environment
%J Kybernetika
%D 1998
%P 411-416
%V 34
%N 4
%U http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a9/
%G en
%F KYB_1998_34_4_a9
Flusser, Jan; Suk, Tomáš. On selecting the best features in a noisy environment. Kybernetika, Tome 34 (1998) no. 4, pp. 411-416. http://geodesic.mathdoc.fr/item/KYB_1998_34_4_a9/

[1] Fukunaga K.: Introduction to Statistical Pattern Recognition. Academic Press, New York 1972 | MR | Zbl

[2] Devijver P. A., Kittler J.: Pattern Recognition: A Statistical Approach. Prentice Hall, London 1982 | MR | Zbl

[3] Abu–Mostafa Y. S., Psaltis D.: Recognitive aspects of moment invariants. IEEE Trans. Pattern Anal. Mach. Intell. 6 (1984), 698–706 | DOI

[4] Teh C. H., Chin R. T.: On image analysis by the methods of moments. IEEE Trans. Pattern Anal. Mach. Intell. 10 (1988), 496–512 | DOI | Zbl

[5] Pawlak M.: On the reconstruction aspects of moment descriptors. IEEE Trans. Inform. Theory 38 (1992), 1698–1708 | DOI | MR | Zbl

[6] Liao S. X., Pawlak M.: On image analysis by moments. IEEE Trans. Pattern Anal. Mach. Intell. 18 (1996), 254–266 | DOI

[7] Flusser J., Suk T.: Invariants for recognition of degraded 1-D digital signals. In: Proc. 13th ICPR, Vienna 1996, vol. II, pp. 389–393

[8] Flusser J., Suk T.: Classification of degraded signals by the method of invariants. Signal Processing 60 (1997), 243–249 | DOI | Zbl