The Side Scan Sonar Target Detection And Noise Suppressing Method Analysis

The side scan sonar system is the main way of underwater target detection, it is the important tool for ocean topography detection, and it is also the highly integration of computer technology, data processing technology and digital sensors and other high-tech, it is on behalf of the most advanced technology in the field of hydrography and cartography. The side scan sonar technology plays important role in the field of mine clearance, anti-submarine, maritime battle field environment construction, and important military training area wire drag survey.With the development of new materials acoustic array, high performance computers and new signal processing technology, the side scan sonar technology has been able to meet most of the underwater target detection accuracy requirements. But due to the various influence factors, such as the work mode, marine environment factors and bottom slope, the relative position of the echo point in sonar image is disproportionate to the relative position of the corresponding seabed echo point, that is to say, the image has distortion problem, which will affect the accuracy of the judgment.The paper aimed at the above question, studied systematical from the side scan sonar data into graph preprocessing, sonar image distortion and correction, echo statistics analysis, noise suppression and edge detection. Through theoretical derivation, mathematical modeling and experimental verification, the signal noise suppression method is given, which can provide technical support for improving the accuracy of sonar image interpretation, ensure the results quality of submarine exploration, and provide data information assurance of high quality for naval maritime battle environmental construction. The main research result as followed:1. The related theory of side scan sonar target detection was introduced systematical, the acoustic principle of side scan sonar was emphasized on, and then the acoustic characteristics and operation indicators were analyzed, the detecting methods for submarine targets were discussed, the problems of indeterminate technical indicators in the current practice were solved.2. Through the decoding scan sonar original data format, the processing which the scan sonar raw data formed image is studied, then a gesture data smoothing method based on the least square method is given. The statistics laws of side scan sonar echo data are analyzed, and the K distribution model can fit echo amplitude histogram is verified. A real-time detection method for loud amplitude signal was put forward, which based on the false-alarm function. The simulation results showed the method can detect the target echo signal in the Ping signal effectively, however, this method did not work in the individual matter when the target echo amplitude less than the bottom materials echo amplitude.3. According to the original data noise of side scan sonar with high frequency, and it also has the characteristics of superior difference with the ambient acoustic amplitude, the advantage of wavelet transform has good localization in both time and frequency domain is used, the threshold processing method of one-dimensional discrete wavelet transform coefficients is put forward, exploiting one-dimensional discrete wavelet transform to denoise Ping signal of side scan sonar. The experimental results show, the method can resolve the singular point problem of side scan sonar Ping amplitude sequence more effective, thus the sonar image noise can be well suppressed, and the sonar image details remain outstanding after denoising.