Clustering Statistic Hough Transform Based Estimation Method for Motion Elements of Multiple Underwater Targets

Estimation for motion elements is one of the core functional components of the multiple targets tracking system. Aiming at estimation for motion elements of multiple underwater targets, a Clustering Statistic Hough Transform (CSHT) method is proposed in order to overcome the false alarm and missing detection effects as well as positioning errors of the sonar data and improve the accuracy and reliability of feature extraction. First, the distance-direction data from the multi-beam forward-looking sonar mounted on the unmanned underwater vehicle are transformed to position curves of multiple targets in the earth-fixed frame, and the position curves appear to be sampling points that form the data space. Second, parameter space is constructed by applying Hough transform to the sampling points in the data space, and then the votes of each cell in the rasterized parameter space are accumulated. Finally, fuzzy iterative self organizing data analysis techniques algorithm clustering method is exploited for extraction of multiple peaks in the parameter space to realize estimation for motion elements. The application of CSHT method in the underwater multiple targets tracking system is further explained in this paper. Simulation results demonstrate that CSHT method is insensitive to environment noise, false alarm and missing detection effects of the sonar and offers favorable estimation accuracy and tracking performance, indicating engineering reliability.