Micro-Motion Behavior Recognition of Target by ISAR Images Based on MDJ-STT

The micro-motion behavior of key components on the target plays an important role in judging its state. In recent years, inverse synthetic aperture radar (ISAR) has become an important means in the field of target detection due to its ability to obtain motion features and structural features. In order to overcome the limitations of single-domain features in recognizing micro-motion behavior, this paper proposes an end-to-end micromotion behavior recognition model MDJ-STT, which adopts a complex-valued (CV) architecture to achieve effective feature extraction of CV ISAR data and time-frequency data, and then effectively fuses the multi-domain features in the dimensions of time, space and frequency of the target. Finally, the attention of the fused feature sequences is optimized, which can achieve effective recognition of the spin, precession and nutation of the target antenna. In the experiment part, the recognition results of the method proposed in this paper are better than other methods on multiple datasets, and the recognition accuracy, recall rate and F1 value are all over 95%.