Target Recognition from HRRP Based on Physical Optics Method

In this paper, the high-resolution range profile (HRRP) of targets is firstly established with the far-field scattered field. And the scattered field is obtained by the physical optics (PO) method. The convolutional and pooling layers of the convolutional neural network (CNN) are extracted local spatial features from the HRRP sequence. The fully connected layers combine the features of each module to complete the extraction and classification recognition of the overall features.