Acoustic Target Classification Using Multiscale Methods

Abstract

This study considers the classification of acoustic signatures using features extracted at multiple scales from hierarchical models and a wavelet transform, In the model based approach; multiscale spectral features are extracted with hierarchical autoregressive and moving average (ARMA) models. The modeling approach is also used for monitoring vehicular activities from an AR spectrogram. The AR spectrogram shows engine speed; gear changes; and other vehicular activities well; because it represents dominant spectral peaks better than a short time Fourier transform. In the wavelet transform based approach; multiscale features are obtained with a wavelet transform. Multiscale classification methods were applied to acoustic data collected at different test tracks under various testing conditions. In this experiment; about 92 percent of vehicles were correctly identified.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1998
Accession Number
ADA358579

Entities

People

  • D. Hillis
  • K. Eom
  • M. Wellman
  • N. Srour
  • R. Chellappa

Organizations

  • United States Army Research Laboratory

Tags

Communities of Interest

  • Energy and Power Technologies
  • Sensors

DTIC Thesaurus Topics

  • Acoustic Signals
  • Acoustic Signatures
  • Background Noise
  • Classification
  • Coefficients
  • Data Sets
  • Diesel Engines
  • Feature Extraction
  • Frequency
  • Frequency Response
  • Ground Vehicles
  • Machine Learning
  • Neural Networks
  • Power Spectra
  • Target Classification
  • Vehicles
  • Wavelet Transforms

Readers

  • Acoustical Oceanography.
  • Computational Fluid Dynamics (CFD)
  • Spectroscopy.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference