Moving target detection in sequential SAR images using kernel correlation method

Moving targets is difficult to detect in synthetic aperture radar (SAR), as its highly dynamic moving makes the SAR image normally defocused. This paper proposes a kernel correlation method to detect moving targets based on the characteristics of sequential SAR complex images. Firstly, the working mode for high frame-rate sequential SAR images and the frequency spectrum characteristics are analyzed. Then, considering the moving target detection method in the optical image, kernel function theory is introduced. Each pixel in the SAR image is classified into two categories according to whether moving targets are present or not. Next, the problem of target detection in sequential SAR images is transformed into one-dimensional temporal disturbance detection, the mechanism of separation of moving targets and clutter based on kernel correlation is analyzed, and the corresponding detection method is presented. The combination of the real signal and the imaginary signal could further improve the moving target detection performance. Finally, simulation experiments on air target detection are carried out, and the results of the experiment verify the effectiveness of the proposed method.