Salt & pepper noise impact on image coding based on image splitting

Abstract

The paper deals with the influence of noise and two control parameters on a segmentation method for image coding. Dynamic min_max criterion and variance criterion were used as segmentation control parameters. For the segmentation of images we have used the method of image splitting. The influence of the value of variance and dynamic criterion on the quality of the segmentation coded image is demonstrated. We have evaluated the objective criterion of the quality of the image (peak signal-to-noise ratio) and number of regions in the image for different values of segmentation control parameters and noise density.