Design of Remote Sensing Image Automatic Target Recognition and Emergency Rescue Assistant System Based on Deep Learning

The purpose of this study is to further improve the accuracy and efficiency of RS (Remote sense) image automatic target recognition through DL (Deep learning) technology, and optimize the design and implementation of emergency rescue assistance system. In this article, an automatic target recognition method of RS image based on DL is proposed, which can realize accurate recognition of different targets. Moreover, an emergency rescue auxiliary system is designed based on this method. The system includes rescue decision support module, resource optimization allocation module and real-time monitoring module, which can provide all-round decision support and optimize resource allocation for rescue. Finally, a large quantity of RS image data are tested and analyzed to verify the feasibility and effectiveness of this study. The results show that the target detection accuracy of the improved algorithm has basically reached more than 90%, which is obviously improved compared with the original algorithm. Moreover, in the test set and experimental set, the response speed of the system in this article is fast, which is about 0.35s s. The results show that the automatic target recognition and emergency rescue auxiliary system based on DL is feasible and effective, and can provide effective technical support for practical applications such as emergency rescue.