An Image Denoising Method Based on BM4D and GAN in 3D Shearlet Domain

Table 2

The results of the different method for standard Gaussian white noise with .

PSNR SSIM EPI Time/s	BM3D	EPLL	TNRD	WNNM	WSNM	Proposed
Cameraman	31.91	31.81	32.18	32.17	32.15	33.01
	0.72	0.71	0.76	0.80	0.79	0.86
	0.39	0.33	0.39	0.44	0.47	0.59
	0.51	29.86	0.88	136.75	172.14	3.51
Barbara	33.69	31.81	32.15	43.26	34.27	34.15
	0.77	0.75	0.77	0.83	0.87	0.86
	0.41	0.40	0.43	0.50	0.59	0.61
	3.05	178.25	1.64	825.50	986.36	7.62
Lena	34.94	32.91	33.44	33.54	33.53	33.94
	0.70	0.77	0.78	0.82	0.88	0.86
	0.40	0.41	0.43	0.48	0.55	0.59
	0.46	30.31	0.90	134.41	165.01	4.90
House	34.94	34.14	34.55	35.15	35.14	35.17
	0.69	0.78	0.81	0.81	0.83	0.87
	0.44	0.43	0.51	0.58	0.58	0.60
	0.78	32.32	0.87	136.16	170.27	4.74
Peppers	32.70	32.58	33.03	32.97	32.97	33.01
	0.66	0.81	0.83	0.84	0.85	0.84
	0.32	0.35	0.42	0.55	0.58	0.58
	0.46	29.55	0.91	132.65	160.62	3.89
Baboon	27.78	27.85	27.89	27.93	27.89	28.18
	0.64	0.77	0.79	0.80	0.84	0.84
	0.28	0.28	0.44	0.50	0.53	0.56
	1.49	111.98	1.22	490.38	654.31	7.11