基于组合特征的水下三维目标检测跟踪算法  

A Combined Feature-based Algorithm of Target Detection and Tracking Used

作　　者：曾腾 任露露 王宇杰[1,2] 王朋 黄海宁[1,2] ZENG Teng;REN Lulu;WANG Yujie;WANG Peng;HUANG Haining(Key Laboratory of Science and Technology on Advanced Underwater Acoustic Signal Processing,Chinese Academy of Science,Beijing 100190,China;Institute of Acoustics,Chinese Academy of Sciences,Beijing 100190,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]中国科学院先进水下信息技术重点实验室,北京100190 [2]中国科学院声学研究所,北京100190 [3]中国科学院大学,北京100049

出　　处：《兵工学报》2023年第5期1384-1393,共10页Acta Armamentarii

摘　　要：针对三维成像声呐水下人工目标检测跟踪困难的问题,提出一种基于组合特征的水下三维目标检测跟踪方法。该算法利用声图的强度和距离等分布信息,根据目标出现前后声图能量的变化,利用背景差分的方法提取核函数,并在二维声图相关匹配的基础上,通过关联由前后帧图像的Hu不变矩、目标质心和目标距离组成的组合几何特征,实现二维向三维映射的人工目标检测与跟踪。湖试数据处理结果显示,新算法的跟踪性能提升了约7.15%,消耗时间相比传统方法减少了约60%,验证了算法的有效性。To deal with the difficulty of underwater artificial target detection and tracking via three-dimensional(3D)imaging sonar,a combined geometric feature-based underwater 3D target detection and tracking method is proposed.The algorithm uses distribution information such as the intensity and depth of the sonar image,and the background difference method to extract the kernel function according to the changes in the energy of the sonar image before and after the target appears.Besides,the detection and tracking of the artificial target using three-dimensional mapping instead of two-dimensional mapping is realized by correlating combined geometric features including the Hu invariant moment,target centroid and target distance of the previous and subsequent frame images on the basis of the correlation matching of the two-dimensional sonar image.The data processing results of the lake test show that the algorithm improves tracking performance by about 7.15%,and consumes about 60%less time than the traditional method,which verifies the effectiveness of the algorithm.

关 键 词：三维成像声呐 检测跟踪 组合几何特征 匹配跟踪 

分 类 号：TN911.7[电子电信—通信与信息系统]