Research Article

Variable Splitting Based Method for Image Restoration with Impulse Plus Gaussian Noise

Table 3

Computational results on salt-and-pepper noise plus Gaussian noise removal by solving model (7).

Blur ā€‰HNW09 ADM2 ADM3 ADMGB

Gaussian30%23/39.08/20.6514/30.72/20.68 45/6.53/20.9546/10.08/20.92
50%29/49.66/20.4115/32.61/20.42 50/7.66/20.5751/11.28/20.59
70%36/61.31/20.0917/36.58/20.10 66/9.95/20.1267/13.55/20.13
90%56/89.73/18.7430/64.38/18.71143/21.48/18.77154/28.22/18.62

Out-of-focus30%19/32.78/22.6613/27.95/22.30 46/6.67/22.5947/10.13/22.48
50%24/40.27/21.9514/30.08/21.72 53/7.91/21.9654/11.20/21.92
70%35/59.88/20.8917/37.61/20.74 85/11.89/20.8294/19.78/20.89
90%57/97.95/18.4333/71.48/18.57144/21.27/18.58146/30.77/18.42