Automatic Classification Algorithm of Underwater Moving Targets near Coast

Lu, D. and Xu, C., 2019. Automatic classification algorithm of underwater moving targets near coast. In: Hoang, A.T. and Aqeel Ashraf, M. (eds.), Research, Monitoring, and Engineering of Coastal, Port, and Marine Systems. Journal of Coastal Research, Special Issue No. 97, pp. 217–222. In order to improve the detection and recognition ability of near-coastal underwater moving targets, a multi-trajectory automatic classification algorithm based on multi-trajectory automatic classification and pixel tracking fusion is proposed. The 3D imaging model of the underwater moving target near the coast is constructed, and the template matching and adaptive segmentation of the underwater moving target image near the coast are carried out by using the similarity trajectory tracking method, and the pixel features of the underwater moving target near the coast are extracted by combining the geometric edge reconstruction method. The pixel information of the underwater moving target image near the coast is extracted by uniform pixel fusion technology, and the edge profile detection and feature search of the underwater moving target near the coast are carried out by combining the gray discretization processing technology, and the automatic classification of multiple tracks in the complex water environment is realized. The simulation results show that the algorithm has better image processing ability, higher image quality and better accuracy of feature extraction, which improves the ability of feature extraction and multi-trajectory automatic classification of underwater moving targets near the coast.