Vehicle Target Detection in Complex Scene SAR Images Based on Co-saliency

Complex features and topographies such as urban areas, forests, and mountains in complex scenes cause many types of SAR image clutter and interference, which makes it more difficult to detect vehicle targets. Aiming at the problem of vehicle target detection in complex scenarios, this paper proposes a SAR image vehicle target detection algorithm based on co-saliency, which uses auxiliary data and data to be measured to calculate co-saliency, obtains the salient features of the target to be measured to complete the detection task, and can effectively detect vehicle targets with the assistance of a small number of vehicle target slices. The co-saliency is applied to the SAR image target detection task, and the preliminary co-saliency map is obtained by calculating the contrast features and corresponding features of the image. Then the weights generated by the co-saliency map of the auxiliary data are used to further suppress the background and enhance the target, so as to improve the detection performance. Experimental verification is performed on the SAR images of the measured complex scenes, and the performance of the proposed method is verified. The experimental results confirm that compared with the traditional CFAR algorithm and significance algorithm, the detection accuracy of the co-saliency algorithm is improved by at least 0.469, and the F1-score is improved by at least 0.561.