HANet: A Hierarchical Attention Network for Change Detection With Bitemporal Very-High-Resolution Remote Sensing Images

Benchmark Model Rank Results
change-detection-on-cdd-dataset-season-1HANet#18F1-Score: 89.23F1: 89.23Precision: 92.86Recall: 85.87
change-detection-on-dsifn-cdHANet#7F1: 62.67Precision: 56.52Recall: 70.33Overall Accuracy: 85.76
change-detection-on-googlegz-cdHANet#4F1: 75.28Precision: 78.58Recall: 72.25Overal Accuracy: 88.34
change-detection-on-levirHANet#9F1: 77.56Prcision: 79.70Recall: 75.53OA: 98.22KC: 76.63
change-detection-on-levir-cdHANet#25F1: 90.28IoU: 82.27Overall Accuracy: 99.02F1-score: 90.28
change-detection-on-s2lookingHANet#10F1-Score: 58.54Precision: 61.38Recall: 55.94OA: 99.04KC: 58.05
change-detection-on-sysu-cdHANet#12F1: 77.41Precision: 78.71Recall: 76.14OA: 89.52KC: 70.59
change-detection-on-whu-cdHANet#20F1: 88.16Precision: 88.30Recall: 88.01Overall Accuracy: 99.16