Share Email Print

Proceedings Paper

Active biosonar systems based on multiscale signal representations and hierarchical neural networks
Author(s): Gordon S. Okimoto; Reid Shizumura; David W. Lemonds
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Signal features based on multiresolution short-time Fourier transforms (STFT) and the Morlet wavelet transform (MWT) have been developed to classify echo returns from targets ensonified by simulated dolphin echolocation clicks. Spectrogram features are obtained at different scales of resolution using analysis windows of different sizes. A method of compressing the highly redundant time-scale representations provided by the MWT has been developed based on multiscale edge analysis (MSEA) of wavelet local maxima. Neural networks are used to evaluate the efficacy of the various feature sets for target recognition. Hierarchical neural networks are used to combine different feature sets for improved classification performance.

Paper Details

Date Published: 4 September 1998
PDF: 8 pages
Proc. SPIE 3392, Detection and Remediation Technologies for Mines and Minelike Targets III, (4 September 1998); doi: 10.1117/12.324204
Show Author Affiliations
Gordon S. Okimoto, ORINCON Corp. (United States)
Reid Shizumura, ORINCON Corp. (United States)
David W. Lemonds, ORINCON Corp. and Univ. of Hawaii/Manoa (United States)

Published in SPIE Proceedings Vol. 3392:
Detection and Remediation Technologies for Mines and Minelike Targets III
Abinash C. Dubey; James F. Harvey; J. Thomas Broach, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?