Intelligent SAR Target Recognition Technologies for High Dynamic Aerial Platforms : Methods, Challenges, and Future Directions

SAR (Synthetic Aperture Radar) has gained widespread use in reconnaissance, monitoring, and other fields due to its unique advantage of all-weather, all-time imaging. With the evolving demands, intelligent SAR target recognition technology for high dynamic aerial platforms has become a research hotspot in this area. This paper provides a review from two perspectives: the traditional SAR ATR (Automatic Target Recognition) three-stage detection process and deep learning methods. First, the research progress in target recognition within traditional SAR ATR methods is reviewed, explaining their limitations in recognition accuracy. Next, the application of deep learning in SAR target detection and recognition is analyzed, emphasizing the advantages of deep learning technologies in automatic feature extraction and end-to-end training, we also demonstrate their potential to significantly improve recognition accuracy and overcome the limitations of traditional methods. Finally, the major challenges and future development directions of intelligent SAR target recognition technologies for high dynamic aerial platforms are discussed. This paper aims to provide comprehensive references and guidance for the research and application of airborne SAR target recognition technology.