Simulation of ISAR Image Based on the Deep Learning and the Electromagnetic Scattering Mechanism

The Inverse Synthetic Aperture Radar (ISAR) image inversion is of great significance in the fields of Earth observation, ocean monitoring and target recognition. However, traditional ISAR imaging algorithms are difficult to meet the requirements of fast imaging due to high computational cost. In order to solve this problem, in this paper, the electromagnetic scattering mechanism and deep learning technology are combined to improve the imaging speed of ISAR and save the computational cost. And the results have been compared with those obtained by the Back Projection (BP) algorithm to show the effectiveness and accuracy of the proposed method. The results show that the proposed DL-based ISAR imaging method improves the imaging speed with the same imaging quality.