Fast-Clustering Method for X-Band Radar Point Cloud Data

X-band pulse-compression radar, as a leading front-end sensing device, plays an important role in maritime surveillance. For the detection of targets within X-band radar point cloud data, clustering algorithms are typically employed for target extension before detection. However, traditional clustering algorithms generally exhibit low performance when processing X-band radar point cloud data in practical applications. In this letter, we propose a clustering algorithm capable of rapidly partitioning X-band radar point cloud data. First, we employ a Pre-sampling method to reduce the time complexity of the algorithm, and then utilize adaptive thresholds and breakpoint check to enhance the accuracy of the algorithm. To validate the effectiveness of the proposed algorithm, we deployed X-band pulse compression radar on the coast of Hainan, China, and collected a substantial amount of measured data. Experimental results demonstrate that our method outperforms traditional clustering algorithms in processing X-band radar point cloud data, achieving an ACC of 96.27% and an NMI of 98.17% in much less time.