一种图像回归与关联关系特征融合的遥感影像变化检测方法

马宗方,郝凡,宋琳,麻瑞

Image Regression and Association-based Feature Fusion for Remote Sensing Image Change Detection

Zongfang MA,Fan HAO,Lin SONG,Rui MA

表2 不同方法在Sardinia数据集的变化检测精度

Table 2 Change detection accuracy of different methods on Sardinia dataset

方法 FN FP OE Ra Rp Rm Rf Ka PCC 3 323 29 550 32 873 0.734 0.127 1 0.435 7 0.872 9 0.118 7 CVA 2 234 43 278 45 512 0.631 8 0.110 8 0.292 9 0.889 2 0.095 IR-MAD 1 740 26 884 28 624 0.768 4 0.179 6 0.228 2 0.820 4 0.212 6 SCCN 6 220 2 078 8 298 0.932 9 0.727 5 0.528 6 0.272 5 0.537 5 FPMSMCD 1 522 3 390 4 912 0.960 3 0.642 9 0.199 6 0.357 1 0.692 SCASC 3 127 3 420 6 547 0.947 0.568 1 0.41 0.431 9 0.550 6 GIR-MRF 1 852 3 291 5 143 0.958 4 0.637 0.242 9 0.363 0.669 7 IRAF 1 578 1 289 2 867 0.976 8 0.824 3 0.206 9 0.175 7 0.796 1