DocumentCode :
3021328
Title :
Multi-modal nonlinear feature reduction for the recognition of handwritten numerals
Author :
Zhang, Peng ; Suen, Ching ; Bui, Tien D.
Author_Institution :
Concordia University
fYear :
2004
fDate :
17-19 May 2004
Firstpage :
393
Lastpage :
400
Abstract :
A novel method of multi-modal nonlinear feature reduction is proposed for the recognition of handwritten numerals. In order to find an effective decision boundary, each class is divided into several clusters. Then the k-NN sorting algorithm is applied to each cluster to get the training data along the effective decision boundary. Optimal discriminant analysis is implemented by multimodal nonlinear mapping to generate a between-class scatter matrix, which requires less CPU time than other nonparametric approaches. Experiments demonstrated that our proposed method could achieve a high feature reduction without sacrificing much discriminant ability. As a result, this new method can reduce ANN training complexity and make the ANN classifier more reliable. Its feature dimensionality reduction outperforms the PCA and mono-modal nonparametric analysis.
Keywords :
Brillouin scattering; Clustering algorithms; Feature extraction; Handwriting recognition; Machine intelligence; Optical character recognition software; Pattern recognition; Principal component analysis; Sorting; Training data;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer and Robot Vision, 2004. Proceedings. First Canadian Conference on
Conference_Location :
London, ON, Canada
Print_ISBN :
0-7695-2127-4
Type :
conf
DOI :
10.1109/CCCRV.2004.1301474
Filename :
1301474
Link To Document :
بازگشت