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

Table 1

Example of regression training vectors (LIBSVM format) taken from images in Figure 3.

4	1 : 255	2 : 255	3 : 8	4 : 255	5 : 255	6 : 0	7 : 4	8 : 8
8	1 : 255	2 : 8	3 : 12	4 : 255	5 : 12	6 : 4	7 : 8	8 : 12
12	1 : 8	2 : 12	3 : 16	4 : 255	5 : 16	6 : 8	7 : 12	8 : 16
16	1 : 12	2 : 16	3 : 20	4 : 12	5 : 20	6 : 12	7 : 16	8 : 255
20	1 : 16	2 : 20	3 : 255	4 : 16	5 : 255	6 : 16	7 : 255	8 : 24