Research And Application Of Synthetic Aperture Sonar Image Preprocessing Technology

Synthetic Aperture Sonar(SAS)is a high-resolution imaging sonar.Through the uniform linear motion,the echo signals of a plurality of small-sized arrays at different positions are coherently accumulated to obtain a large-sized virtual aperture.The signal frequency is independent of the distance and the azimuth is high resolution.However,the SAS raw image is heavily affected by noise and the image must be further processed.At the same time,the display and control software is an important software running in the SAS imaging system,responsible for image pre-processing,real-time display and control functions.Designing a high-performance display and control software can help improve the overall system performance.The SAS is equipped on the unmanned Underwater Vehicle(UUV)for intelligent underwater operation,and the real-time SAS image is obtained by using the real-time display control of the display and control software.It is a research hotspot in the field of sonar.This paper focuses on the preprocessing techniques of SAS raw images,including image equalization,motion compensation,slant range correction,denoising algorithm,and pseudo color processing.Designed and implemented a display and control software based on UUVSAS system.Firstly,this paper studies the basic principles of SAS imaging,and fully analyzes the factors affecting the quality of SAS images,and studies the background noise model.For the above analysis,two algorithms for image equalization are studied.Phase gradient autofocus algorithm,slant range correction algorithm and pseudo color processing algorithm are studied.Several denoising algorithms are analyzed,including classical image filtering methods,Lee algorithm and Kuan algorithm.Additionally compare the performance of various algorithms,selected for the display and control of software system.The motion compensation algorithm phase gradient autofocus is mainly used for error compensation of the entire image,and cannot be directly used in the display and control software of the real-time imaging system.The system adopts the method of first dividing the image,calculating one by one and then splicing,and realizing the error compensation for the real-time image.Normal sonar pseudo-color processing is directly designed using the RGB color space,without considering the eye's sensitivity to color.The system adopts the method of first setting the transformation curve in the RGB color space,then transforming to the HSV space which is more closely matched with the human eye,and finally transforming back to the RGB space to generate the palette.The ability to discriminate artificially on SAS images is enhanced.Then,the UUVSAS system-based display and control software is designed.The basic image information real-time display,control,partial enlargement,and multi-pseudo color matching functions are realized,and the track map display function of real-time display UUV position and scanning area is proposed.Finally,the experimental SAS raw data was used to verify the results.The improved image preprocessing algorithm achieved the desired results.The functions of the display and control software have been achieved and the performance is good.Additionally the research content of this system is summarized,and the problems that need further improvement and research are put forward.