Extraction and classification of acoustic scattering from underwater target based on Wigner-Ville distribution

摘要

The classification of rigid and elastic scattering is important for target shape and size estimation in underwater target recognition. A Wiper-Ville distribution (WVD)-based method for target scattering extraction and classification is proposed by taking the highlight model as the theoretical basis. Firstly, the WVD matrix of the target echo is rotated and filtered in the transform domain to remove cross-terms. Secondly, the instantaneous frequency characteristics are utilised for highlight detection and positioning according to the difference between target scattering and background (noise and reverberation) in the transform domain. Finally, the spectral peak frequency, half-power bandwidth and spectral centroid are selected for highlight classification due to the mid frequency enhancement of elastic scattering. Experimental result shows a rate of 98.9% for recognising rigid and elastic scattering.