Image Demosaicing Based on Generative Adversarial Network

Table 2

CPSNR and SSIM of the test results for the Kodak database by different methods.

No.	1		2		3		4		5		6		7		8
	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM
Bilinear	33.55	0.8060	38.66	0.8984	39.59	0.9333	38.56	0.9118	34.22	0.8777	34.94	0.8443	39.16	0.9538	33.19	0.8243
TL	32.98	0.8081	36.91	0.8564	38.64	0.9201	37.16	0.8788	33.76	0.8500	34.14	0.8364	37.06	0.9288	32.75	0.8323
HQL	36.02	0.9505	41.39	0.9602	43.13	0.9761	42.32	0.9709	37.30	0.9701	37.40	0.9566	42.47	0.9827	35.17	0.9502
Zhang’s	38.79	0.9881	40.46	0.9754	41.45	0.9861	40.11	0.9793	37.05	0.9859	40.23	0.9886	40.25	0.9877	35.99	0.9865
LDI-NAT	33.69	0.9633	39.01	0.9683	40.40	0.9802	39.02	0.9718	34.77	0.9797	34.90	0.9693	40.93	0.9869	32.01	0.9676
ARI	38.84	0.9863	39.69	0.9704	42.75	0.9874	40.66	0.0904	38.29	0.9887	40.55	0.9885	42.74	0.9888	34.99	0.9801
MLRI	36.86	0.9818	41.49	0.9807	42.95	0.9883	41.99	0.9842	37.68	0.9885	38.99	0.9865	42.79	0.9904	35.17	0.9825
RI-GBTF	39.03	0.9880	40.08	0.9753	41.51	0.9870	40.49	0.9803	37.81	0.9878	40.89	0.9899	40.93	0.9869	32.01	0.9676
DnCNN	41.96	0.9874	43.73	0.9732	42.18	0.9869	43.26	0.9867	37.77	0.9849	40.26	0.9873	42.08	0.9873	36.49	0.9738
Proposed	42.07	0.9882	43.83	0.9757	42.22	0.9876	43.93	0.9880	39.49	0.9897	41.77	0.9901	43.53	0.9894	36.88	0.9869
No.	9		10		11		12		13		14		15		16
	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM
Bilinear	38.58	0.9198	38.82	0.9191	36.02	0.8680	38.89	0.9063	32.51	0.7690	35.54	0.8709	38.76	0.9174	36.88	0.8748
TL	38.18	0.9156	38.07	0.9083	35.34	0.8545	37.80	0.8975	32.32	0.7867	34.58	0.8467	37.97	0.9058	35.80	0.8601
HQL	41.59	0.9702	42.41	0.9722	38.75	0.9611	42.41	0.9697	35.03	0.9420	38.54	0.9616	41.74	0.9659	39.87	0.9629
Zhang’s	42.27	0.9848	42.04	0.9858	39.85	0.9868	42.83	0.9868	35.01	0.9821	35.52	0.9806	39.12	0.9758	43.73	0.9900
LDI-NAT	40.11	0.9744	39.81	0.9778	36.13	0.9728	40.72	0.9777	29.64	0.9461	35.41	0.9718	37.58	0.9671	38.56	0.9746
ARI	41.96	0.9714	41.67	0.9819	39.67	0.9843	43.25	0.9869	35.22	0.9804	37.57	0.9836	38.64	0.9715	43.33	0.9892
MLRI	42.36	0.9817	42.44	0.9847	39.36	0.9863	43.37	0.9868	33.15	0.9732	37.75	0.9834	39.85	0.9777	42.84	0.9885
RI-GBTF	42.54	0.9848	42.61	0.9866	40.24	0.9880	43.39	0.9876	35.32	0.9824	36.31	0.9817	38.90	0.9738	44.38	0.9901
DnCNN	42.76	0.9857	40.04	0.9836	42.32	0.9875	42.15	0.9870	36.94	0.9837	39.50	0.9803	40.95	0.9725	44.65	0.9913
Proposed	43.43	0.9883	43.60	0.9871	42.66	0.9901	43.99	0.9894	37.03	0.9861	40.02	0.9845	42.36	0.9779	44.96	0.9904
No.	17		18		19		20		21		22		23		24
	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM	CPSNR	SSIM
Bilinear	38.58	0.9244	35.55	0.8730	36.01	0.8750	38.53	0.9198	35.81	0.8878	36.65	0.8816	41.32	0.9551	35.53	0.8774
TL	37.82	0.9136	34.84	0.8595	35.46	0.8798	38.04	0.9132	35.24	0.8720	36.01	0.8779	40.03	0.9399	35.24	0.8790
HQL	42.17	0.9760	38.50	0.9612	38.20	0.9619	41.27	0.9651	38.42	0.9640	37.78	0.9606	44.46	0.9793	38.54	0.9663
Zhang’s	41.67	0.9882	37.13	0.9794	40.68	0.9856	40.94	0.9769	39.12	0.9824	37.96	0.9741	41.99	0.9820	35.38	0.9853
LDI-NAT	38.31	0.9789	33.89	0.9653	37.22	0.9692	38.31	0.9672	35.22	0.9686	37.11	0.9719	41.80	0.9811	31.77	0.9731
ARI	41.32	0.9868	36.91	0.9753	40.65	0.9821	40.89	0.9760	39.28	0.9748	38.08	0.9728	43.24	0.9814	35.36	0.9849
MLRI	40.69	0.9865	36.55	0.9782	40.03	0.9823	40.71	0.9770	38.17	0.9802	38.55	0.9772	43.67	0.9857	34.64	0.9842
RI-GBTF	41.75	0.9890	37.38	0.9805	41.12	0.9860	41.21	0.9780	39.63	0.9833	38.36	0.9763	41.96	0.9816	35.01	0.9858
DnCNN	42.18	0.9849	38.96	0.9815	42.74	0.9847	40.55	0.9873	40.08	0.9825	39.79	0.9805	44.67	0.9843	39.13	0.9867
Proposed	42.25	0.9891	39.47	0.9983	42.96	0.9926	42.86	0.9877	40.87	0.9894	40.37	0.9862	44.98	0.9916	40.07	0.9954