Space Objects Recognition Based on ISAR Image

In this paper, a method of spatial object recognition based on ISAR image is studied, and the recognition result of four kinds of spatial objects is given by simulation. Firstly, the feature of ISAR image is extracted based on two-dimensional wavelet transform and Gabor wavelet transform, and then the sub-optimal search method is used to compress the feature to improve the efficiency. Finally, the BP neural network is used to identify the target. The simulation result validates the effectiveness of the proposed algorithm.