Wavelet-based feature extraction methods for classification applications

This work discusses feature extraction and dimension-reduction issues in the context of classification applications. A new type of projection pursuit algorithm is presented. This scheme is designed to reduce the number of class features obtained from the wavelet packet decomposition of the signals to be classified. Results show the simple scheme can be used to classify various types of signals in a noisy environment.

[1]  Alan S. Willsky,et al.  A Wavelet Packet Approach to Transient Signal Classification , 1995 .

[2]  Danny Coomans,et al.  Classification Using Adaptive Wavelets for Feature Extraction , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  J. Friedman.  Exploratory Projection Pursuit , 1987 .

[4]  M.P. Fargues,et al.  Comparing wavelet transforms and AR modeling as feature extraction tools for underwater signal classification , 1995, Conference Record of The Twenty-Ninth Asilomar Conference on Signals, Systems and Computers.

[5]  Nathan Intrator,et al.  BCM theory of visual cortical plasticity , 1998 .

[6]  R. Coifman,et al.  Local feature extraction and its applications using a library of bases , 1994 .

[7]  Martin T. Hagan,et al.  Neural network design , 1995 .

[8]  Rama Chellappa,et al.  Dimensionality reduction of multi-scale feature spaces using a separability criterion , 1995, 1995 International Conference on Acoustics, Speech, and Signal Processing.

[9]  Nathan Intrator,et al.  Objective function formulation of the BCM theory of visual cortical plasticity: Statistical connections, stability conditions , 1992, Neural Networks.

[10]  Nathan Intrator,et al.  Classification of underwater mammals using feature extraction based on time-frequency analysis and BCM theory , 1998, IEEE Trans. Signal Process..

[11]  Ronald R. Coifman,et al.  IMPROVED LOCAL DISCRIMINANT BASES USING EMPIRICAL PROBABILITY DENSITY ESTIMATION , 1996 .

[12]  M. Victor Wickerhauser,et al.  Adapted wavelet analysis from theory to software , 1994 .