Feature Extraction of Underwater Target Ultrasonic Echo Based on Wavelet Transform

Abstract

Ultrasonic detection has been widely used in underwater detectoscopes as an important method for underwater detection. Feature extraction of echo signal time-delay and amplitude is the main task of processing underwater ultrasonic signal. Underwater target ultrasonic echo signal is influenced by reverberation and noise from the sea and system itself, reverberation interference of signal background is the main difficulty for target echo detection. So we use denoising algorithm to denoise echo signal. At first this paper denoises the measured weighted background clutter data using wavelet threshold denoising method, then the paper extracts breaking points of echo signal through wavelet transform, at last the paper makes an envelope extraction using Hilbert transform combined with wavelet transform methods, and acquires the feature information of echo signal amplitude.

A Novel Visual Measurement Method for Three-Dimensional Trajectory of Underwater Moving Objects Based on Deep Learning

“…For underwater equipment installation on different occasions, people pay more and more attention to its stability and high precision, and this is an urgent problem that needs to be solved for marine engineering equipment. Fortunately, the detection direction of underwater targets has been greatly developed, for example using methods based on acoustic sensors [1][2][3][4], lidar [5][6][7][8], geometric scattered waves [9][10][11], sonar pictures [12][13][14][15], hybrid particle filter tracking [16], wavelet transform [17][18] and others to detect underwater targets.…”