Automatic classification of positive time-frequency distributions

Author

Pitton, James W. ; Atlas, Les E.

Author_Institution

Dept. of Electr. Eng., Washington Univ., Seattle, WA, USA

fYear

1994

fDate

25-28 Oct 1994

Firstpage

361

Lastpage

364

Abstract

A method of performing automatic classification of positive time-frequency distributions is presented. These distributions are computed via constrained optimization, minimizing the cross-entropy of the distribution subject to a set of constraints. An algorithm for clustering using cross-entropy as the distance measure between vectors was derived by Shore and Gray (see IEEE Trans. PAMI, vol.4, no.1, p.11-17). We apply this method to the time-frequency case, and derive an efficient classification scheme. An advantage of this method is that the time-frequency distributions of the data to be classified do not need to be directly computed; thus, the method can be applied to real-time classification

Keywords

entropy; optimisation; pattern classification; signal representation; signal synthesis; spectral analysis; statistical analysis; time-frequency analysis; automatic classification; clustering algorithm; constrained optimization; cross-entropy minimisation; distance measure; pattern classification; positive time-frequency distributions; real-time classification; signal representation; signal synthesis; spectral analysis; speech recognition; vectors; Clustering algorithms; Constraint optimization; Distributed computing; Interactive systems; Laboratories; Signal resolution; Signal synthesis; Spectrogram; Speech; Time frequency analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Time-Frequency and Time-Scale Analysis, 1994., Proceedings of the IEEE-SP International Symposium on

Conference_Location

Philadelphia, PA

Print_ISBN

0-7803-2127-8

Type

conf

DOI

10.1109/TFSA.1994.467336

Filename

467336