The Scientific World Journal / 2014 / Article / Tab 3

Research Article

A “Salt and Pepper” Noise Reduction Scheme for Digital Images Based on Support Vector Machines Classification and Regression

Table 3

Table results in mean PSNR for different impulse noise reduction methods. The mean is obtained for different noise ratios in 10 test images. Total column reflects the total mean for all ratios and images.

MethodNoise percentage
10 20 30 40 50 60 70 80 90 Total

ACWM [19]31.2427.4922.8218.4814.8912.109.787.946.4716.80
AMF [12]32.1730.4528.7727.1825.5923.9721.8718.0312.0324.45
DBAIN [20]36.5832.9730.5128.5026.6124.8022.9020.8518.0526.86
DSFIRE [21]34.1430.1325.6921.1017.1413.9311.309.237.6018.92
FIDRM [22]36.2832.5330.1228.2926.7625.3923.8922.0819.6827.23
MMEM [14] 27.7730.4629.5228.7127.8426.88 25.76 24.45 22.59 27.11
NAFSM [23]34.2631.3129.5428.1727.0526.0325.0123.8121.2227.38
PSMF [17]26.3525.8425.0823.9721.9518.9114.989.677.6119.37
SDROM [5]30.9228.9327.0725.1922.9720.5317.6214.3410.7922.04
MDBUTMF [24] 37.22 33.86 31.62 29.82 27.94 25.5322.4518.8315.0526.92
SVM-M1 (32-40) 35.4931.9629.3226.5324.6922.3319.7613.075.9623.23
SVM-M1 (64-40) 35.5732.1529.4926.8725.1022.8020.1213.285.9723.48
SVM-M2 (32-40) 36.0232.9630.8629.1927.7426.4225.1423.8022.0928.25
SVM-M2 (64-40) 35.9632.9030.8129.1427.6926.3725.1123.7722.0628.20
SVM-R (32-40) 35.6332.7130.7629.1427.7026.2524.7122.7119.6327.69
SVM-R (64-40) 35.4732.5930.6829.1027.6726.2324.7022.7119.6327.64