Remote Sensing Image Change Detection Network Based on Twin High-Resolution Representation

Table 1

Results of the ablation experiments of the modules on the CDD dataset and DSIFN dataset.

Model	CDD				DSIFN
	Precision (%)	Recall (%)	F1 (%)	OA (%)	Precision (%)	Recall (%)	F1 (%)	OA (%)
Baseline (vgg_16)	0.9066	0.7554	0.8083	0.9554	0.8331	0.7766	0.7956	0.9276
Baseline (HRNet)	0.9484	0.8577	0.9007	0.9603	0.8643	0.8355	0.8496	0.9436
HRNet + CAM	0.9555	0.8659	0.9085	0.9673	0.9002	0.8443	0.8714	0.9484
HRNet + CAM + SOME	0.9670	0.8798	0.9213	0.9788	0.9104	0.8893	0.8987	0.9537

The best ones are marked in bold.