Two Ways to Improve Average and Median Filters

Abstract

Colored digital images are exposed to various types of noise, the most important of which is salt and pepper noise, which negatively affects the clarity of the image in addition to the negative effects on its characteristics, which calls for providing an effective method that mitigates the negative effects of noise as much as possible. The two filters average and median are considered two of the best filters used to remove salt and pepper noise or mitigate the negative effects of this noise, especially if the noise density is low. However, when this density rises, these filters become ineffective in dealing with the noise. In this paper research we will introduce two ways of improving the operations of average and median filters, these improved filters will be tested and compared with the standard average and median filters, it will be shown how the improvements will enhance the value of the quality parameters MSE and PSNR, some recommendation will be raised to show when and how to use each of the improvement ways.