Marine Weak Moving Target Detection Using Sparse Learning Dictionary

Due to the strong interference of sea clutter, it is difficult to effectively extract interested targets from received signals by traditional detection methods, which limits the target detection rate and computational speed. In this paper, a sea clutter suppression and target signal reconstruction method based on sparse learning dictionary and orthogonal matching pursuit algorithm (OMP) was proposed to improve the target detection rate. This method, based on the sparse theory, adopts the optimized K-Singular Value Decomposition (K-SVD) algorithm to train the overcomplete basis of sea clutter, obtaining the learning dictionary, which is then used for sparse decomposition and reconstruction of interested signals so as to filter sea clutter and extract the main component of the target at the same time. It solves the problems of low matching between traditional fixed dictionary and received signals and poor effect of signal extraction. The experimental results show that the algorithm can effectively detect weak moving targets under high sea conditions.