Target Detection Within Sea Clutter Based on Combined Time-Frequency Characteristics

Abstract

To conquer the disadvantages of conventional target detection based on a statistical distribution, this paper focuses on the combined time-frequency characteristics of sea clutter and its application in the field of target detection. Due to the complex characteristics of sea clutter, it is difficult to detect targets based on a single feature parameter. Therefore, this paper focused on the combined time-frequency characteristics of sea clutter, where the normalized energy feature of third-order Intrinsic Mode Function (IMF3) and Tsallis Entropy (TE) of spectrum are regarded as two-dimensional characteristics for target detector. Then, the combined time-frequency characteristics of clutter and target are analyzed in detail, and Support Vector Machines (SVM) is taken to train the two-dimensional feature parameters. Finally, through research of X-band real sea clutter datasets, the proposed method improves the target detection accuracy compared with the conventional CFAR method.