Abstract: Because the SAR image has rich texture and a lot of noise, the traditional SURF operator is not ideal for the detection of target interest points on the SAR(Synthetic Aperture Radar) image, and it has poor adaptability to interest point detection and a large number of useless feature points, so that the target matching success rate decreases, It proposes a new SAR image matching algorithm which integrates CFAR with SURF. The mixed Gauss model CFAR was  used for the detection of target interest regions, and the SURF operator was used to detect feature extraction. Finally, the improved multilayer elimination method was used for matching feature points. The simulation results show that the proposed method has a high matching rate and good robustness when the scale, rotation and noise vary.

Key words: synthetic aperture radar(SAR) image; target matching; constant false alarm rate(CFAR); SURF algorithm