Title :
Vector quantization technique for nonparametric classifier design
Author :
Xie, Qiaobing ; Laszlo, Charles A. ; Ward, Rabab K.
Author_Institution :
Dept. of Electr. Eng., British Columbia Univ., Vancouver, BC, Canada
fDate :
12/1/1993 12:00:00 AM
Abstract :
An effective data reduction technique based on vector quantization is introduced for nonparametric classifier design. Two new nonparametric classifiers are developed, and their performance is evaluated using various examples. The new methods maintain a classification accuracy that is competitive with that of classical methods but, at the same time, yields very high data reduction rates
Keywords :
approximation theory; data reduction; pattern recognition; vector quantisation; Parzen kernel classifier; condensing algorithm; data reduction rates; design; k-nearest neighbour; nonparametric classifier; vector quantization; Application software; Color; Image analysis; Kernel; Multispectral imaging; Optimized production technology; Pattern analysis; Shape; Spatial resolution; Vector quantization;
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
