Review of ship target detection based on SAR images

Ship target recognition is one of the important directions of the development of intelligent war, which has been widely used in military fields such as early warning detection, precision guidance, command and control, and intelligence reconnaissance. As a consequence, it has always been the research focus of governments and armies. With the advantage of obtaining high-resolution images of targets without interference from weather and lighting conditions, SAR is often applied in the field of ship target detection. The traditional SAR image detection method relying on artificial design features is susceptible to complex background interference. It has the disadvantages of low recognition accuracy, low recognition efficiency and poor generalization ability. The deep learning method gets rid of the dependence on design features and modeling by the ability to autonomously learn parameters and automatically extract features. Compared with the traditional method, it has the advantages of strong anti-interference, stable system structure, strong adaptability and high detection accuracy. Especially, it has great development potential in the field of SAR ship target detection with complex scenes and different target attitudes. Therefore, using deep learning technology to solve the problem of ship target detection is one of the important research directions of intelligent warfare. Meanwhile, reasonable reference to related technologies in other fields can optimize the ability of convolution feature extraction, improve the detection accuracy and performance of algorithm, that is helpful to detect the target more accurately. The fusion of multiple algorithms will provide more channels for SAR image target detection. In this paper, the current SAR image ship target detection methods are summarized and the next research direction is discussed.