Automatic Target Recognition Based On High-resolution Radar Range Images

This thesis mainly studies the problem of automatic target recognition (ATR) based on high range resolution radar profiles. First, the thesis gives a survey on the ATR techniques and makes a brief introduction to the fundamental of ATR. Second, a novel method for joint alignment and ATR, which is based on high range resolution profiles, is presented. Then the new method is successfully extended to the cases of multiple range profiles to reduce the aspect sensibility and improve the ATR performance. Experimental results obtained using the MSTAR and XPATCH database show that the performance of our new approach is better than that of the baseline method of American AFRL's SHARP in the same dataset.