New Regularization Models for Image Denoising with a Spatially Dependent Regularization Parameter

Table 1

Summarized results of the comparative experiment.

	Figure 2(a)	Figure 2(b)	Figure 2(c)	Figure 2(d)
ReErr
Noisy image	0.1064	0.1847	0.1913	0.1643
ROF model	0.0522	0.1279	0.0784	0.0925
LLT model	0.0633	0.1273	0.0807	0.0894
Hybrid model	0.0509	0.1224	0.0758	0.0879
Model 1	0.0361	0.1151	0.0751	0.0842
Model 2	0.0298	0.1132	0.0729	0.0827
SNR
Noisy image	9.00	6.43	5.49	6.42
ROF model	15.19	9.62	13.24	11.41
LLT model	13.51	9.66	12.99	11.71
Hybrid model	15.41	10.00	13.53	11.85
Model 1	18.39	10.53	13.61	12.23
Model 2	20.06	10.68	13.87	12.38
Time
ROF model	0.28	1.22	1.26	1.05
LLT model	1.33	4.63	1.73	3.06
Hybrid model	2.23	8.44	4.48	5.46
Model 1	2.87	6.36	5.18	4.65
Model 2	2.50	9.63	5.93	6.08