Marine Multi-Target Detection Based on Improved Wavelet Transform

In order to improve the accuracy of ballistic missile against ships, it is of great significance to study and analyze the sea targets detection of missile-borne SAR. Aiming at the problem that the detection algorithm based on multi-scale wavelet transform is difficult to accurately realize multi-target detection and recognition in the sea under coherent noise, a multi-target detection model based on improved wavelet transform is proposed. Firstly, the traditional wavelet transform based on gradient mode value is improved by using the directional adjustable wavelet function, and a multi-scale wavelet detection model with adjustable direction is constructed to roughly extract the target edge. Then, aiming at the problem that the distinction in the image is not obvious and there are a lot of sharp points, the combination processing of opening operation, Top-hat and Bot-hat filtering and small radius closed operation is used to enhance the distinction. On this basis, aiming at the porous phenomenon existing in the target, we construct a porous filling model based on erosion and complementary intersection to complete the hole filling quickly and realize multi-target edge detection. Finally, the edge detections of islands and ships are compared with three common algorithms to verify the superiority of this model.