Quadratic boundaries in N-N classifiers with dissimilarity-based representations

Author

Horikawa, Yo

Author_Institution

Fac. of Eng., Kagawa Univ., Takamatsu, Japan

Volume

2

fYear

2002

fDate

26-30 Aug. 2002

Firstpage

1039

Abstract

The performance of the nearest neighbor (N-N) classifiers with the dissimilarity-based representations is studied. The dissimilarity-based representations cause quadratic decision boundaries instead of usual piecewise linear boundaries. The correct classification ratio of the N-N classifiers is improved with the use of the dissimilarity-based representations when variations within class differ from each other and the dimensionality of patterns is high.

Keywords

feature extraction; image classification; image representation; image texture; piecewise linear techniques; spectral analysis; N-N classifiers; bispectrum-based features; correct classification ratio; dissimilarity-based pattern recognition; dissimilarity-based representations; nearest neighbor classifiers; piecewise linear boundaries; quadratic decision boundaries; relational discriminant analysis; statistical pattern recognition; texture images classification; Amino acids; DNA; Euclidean distance; Nearest neighbor searches; Pattern analysis; Pattern recognition; Piecewise linear techniques; Prototypes; Sequences;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing, 2002 6th International Conference on

Print_ISBN

0-7803-7488-6

Type

conf

DOI

10.1109/ICOSP.2002.1179966

Filename

1179966