Abstract: Deep learning technology is more and more widely used in the field of SAR （Synthetic Aperture Radar） ship target detection. However， the limited resources of mobile platforms （airborne/spaceborne） limit the application of SAR ship target detection technology based on deep learning. In order to promote the application of deep learning technology on mobile platforms， this paper carried out research on the implementation of Jetson TX2 embedded platform based on the improved YOLOv2 SAR target detection algorithm. A real-time SAR ship target detection system based on two Jetson TX2 embedded platforms is designed. The two Jetson TX2 embedded platforms work together to effectively improve the processing capacity of the system. At the same time， the Ethernet data transmission method is adopted to ensure that the host computer and the High-speed data interaction of Jetson TX2 embedded platform.