Research Article

An Efficient Universal Noise Removal Algorithm Combining Spatial Gradient and Impulse Statistic

Table 4

Comparative restoration results in PSNR (dB) and MAE (the second row) for Gaussian noise.

MethodLena imageBoats image

Noisy image28.1322.1416.3728.1322.1716.40
4.0027.96115.503.9837.91415.23
3 × 3 median filter32.1328.4423.4430.2527.4323.07
2.2673.7026.7442.7344.0867.059
5 × 5 median filter30.4428.9425.9527.6726.7224.63
2.2803.0834.7873.2133.9065.502
Gaussian filter33.1430.1227.0832.0428.8325.86
2.2162.8564.4592.8853.4755.133
Bilateral filter33.9030.6727.1932.6929.2225.94
1.9752.7854.2932.3223.3585.113
ROAD-TRIF33.9730.5727.1132.9229.2325.93
1.8472.8114.2392.1803.3595.111
SG-BF33.6830.6827.3232.4429.1926.03
2.0042.7644.1172.3223.3915.023