基于SFS方法的侧扫声呐图像三维重构  

Three-dimensional reconstruction of side scan sonar imagebased on SFS method

作　　者：刘浩林 覃珊珊 李鹏鸽 张晓晖[1] 刘青[1] LIU Hao-lin;QIN Shan-shan;LI Peng-ge;ZHANG Xiao-hui;LIU Qing(School of Automation and Information Engineering,Xi'an University of Technology,Xi'an 710048,China)

机构地区：[1]西安理工大学自动化与信息工程学院,陕西西安710048

出　　处：《舰船科学技术》2021年第8期125-130,共6页Ship Science and Technology

基　　金：国家自然科学基金资助项目(No.51875447);陕西省自然科学基础研究计划(2021JLM-58);陕西省重点研发计划(No.S2018-YFZDGY-0583)。

摘　　要：针对侧扫声呐受其测量机理和作业模式影响无法直接获取海底表面地形及水下目标物高度信息的问题,提出改进的明暗恢复形状最小化方法。通过从侧扫声呐原始数据文件出发,解析得到斜距量程、采样精度等重要参数以及二维瀑布图。在瀑布图基础上,提出一种基于小波变换的最小化算法,求解得到对应二维图中每一点的相对高度值,从而实现了海底表面地形的高精度三维重构。通过对比实验证明,与2种经典的最小化法相比,基于小波变换的最小化法性能明显提高。Aiming at the problem that the side scan sonar cannot directly obtain the information of the seafloor surface topography and the height of the underwater target due to its measurement mechanism and operation mode,an improved method of minimizing the shape of the light and dark recovery is proposed.Starting from the original data file of the side scan sonar,the important parameters such as the slant range,sampling accuracy,and two-dimensional waterfall diagram are obtained by analysis.Based on the waterfall diagram,a minimization algorithm based on wavelet transform is proposed to obtain the corresponding two-dimensional The relative height value of each point in the figure realizes the high-precision three-dimensional reconstruction of the seabed surface terrain.And through comparative experiments,it is proved that compared with the two classical minimization methods,the performance of the minimization method based on wavelet transform is obviously improved.

关 键 词：侧扫声呐 格式解析 明暗恢复形状 最小化法 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]