Speckle Noise Reduction via Nonconvex High Total Variation Approach

Table 1

Quantitative denoising comparison among the proposed method (), the TGVSNR method, and the NRSNR method. The best values of SNR (db) of each row are displayed in bold.

Image		Noise index	Proposed method	TGVSNR	NRSNR
		SNR	SNR	SNR	SNR
House	3		12.15	11.18	10.80
	9		14.65	13.92	13.74
	15		15.81	15.25	15.17
	21		16.79	16.15	16.04
Livingroom	3		7.63	7.28	6.65
	9		9.72	9.43	9.05
	15		10.75	10.53	10.30
	21		11.46	11.29	11.06
Lena	3		10.65	10.13	9.72
	9		13.35	12.73	12.48
	15		14.46	14.01	13.82
	21		15.44	14.89	14.69
Cameraman	3		11.27	11.05	11.16
	9		13.87	13.39	13.71
	15		15.02	14.60	14.91
	21		15.86	15.40	15.75
Boat	3		8.44	8.12	7.80
	9		10.38	10.11	10.25
	15		11.60	11.47	11.56
	21		12.40	12.24	12.40
Nimes	3		4.37	3.96	4.02
	9		6.80	6.26	6.26
	15		8.03	7.56	7.53
	21		9.00	8.45	8.42
Phantom	3		12.38	10.72	12.08
	9		16.67	14.51	16.09
	15		18.76	16.63	18.28
	21		20.41	17.92	19.61
Fields	3		10.38	10.08	9.50
	9		12.07	11.90	11.56
	15		13.06	12.89	12.68
	21		13.85	13.56	13.41
Peppers	3		11.46	10.90	10.07
	9		13.96	13.57	13.01
	15		15.05	14.86	14.40
	21		16.03	15.69	15.21
Rem1	3		8.18	7.97	7.91
	9		10.47	10.22	10.37
	15		11.70	11.47	11.65
	21		12.53	12.30	12.47