A Novel Faster RCNN with ODN-Based Rain Removal Technique

Table 1

Quantitative result analysis of existing with proposed methods in terms of various measures.

Test images	Method	PSNR	SSIM	FSIM	UIQI
Image 1	DID	22.409	0.889	0.927	0.967
	DSC	22.302	0.863	0.926	0.953
	LP	22.227	0.889	0.919	0.911
	UGSM	22.763	0.909	0.933	0.911
	TIP	18.045	0.836	0.911	0.830
	CVPR	23.325	0.902	0.921	0.875
	KGCNN	27.851	0.957	0.957	0.978
	DDCN	30.276	0.963	0.963	0.979
	FRCNN-ODN	32.899	0.974	0.972	0.981
Image 2	DID	25.294	0.851	0.924	0.813
	DSC	22.219	0.598	0.835	0.904
	LP	23.034	0.780	0.872	0.806
	UGSM	20.626	0.641	0.824	0.784
	TIP	19.215	0.648	0.851	0.757
	CVPR	23.781	0.674	0.882	0.846
	KGCNN	31.231	0.957	0.970	0.930
	DDCN	33.750	0.962	0.962	0.978
	FRCNN-ODN	34.853	0.971	0.975	0.982
Image 3	DID	22.359	0.865	0.902	0.975
	DSC	22.519	0.846	0.898	0.992
	LP	22.393	0.894	0.907	0.987
	UGSM	22.382	0.913	0.921	0.985
	TIP	18.793	0.872	0.903	0.969
	CVPR	24.013	0.902	0.908	0.994
	KGCNN	28.031	0.966	0.954	0.997
	DDCN	33.451	0.964	0.964	0.997
	FRCNN-ODN	34.860	0.973	0.973	0.998