A Fast ISAR Image Prediction Method for Multiple Isomorphic Targets Based on Deep Learning

In order to improve the computational efficiency of dynamic isomorphic multi-target radar imaging, this paper proposes a rapid multi-target inverse synthetic aperture radar (ISAR) imaging method based on prior information from radar parameters. By using an ensemble deep learning model, including a fully connected neural network (FCNN) and a U-Net to train with the simulated electromagnetic field data and the distance between targets, the proposed network has successfully predicted the ISAR image of two targets with an extremely high Structure Similarity Index Measure (SSIM) of 0.961 compared to the images generated by simulation results. This method demonstrates the potential of the ensemble deep-learning network for fast prediction of ISAR images. Numerical and graphical results are presented to evaluate the effectiveness of the efficient imaging method, which indicates that the proposed method is effective for the rapid prediction of multi-target ISAR images.