High-Density Salt and Pepper Noise Removal from Colour Images by Introducing New Enhanced Filter

摘要

Any image captured by a camera or scanner contains some unwanted particles, which are known as noise or salt and pepper. They are black and white color dots that are present in some pixels. The value of those pixels is either 0 or 255. There exist different predefined methods to remove salt and pepper or noise from an image in Matlab. This paper presents an effective salt and pepper filtering and removal algorithm, which has accuracy greater than some of the predefined filters in Matlab like Median Filter, Adaptive Median Filter, MDBUTMF, MDBPTGMF, and Adaptive Weighted Median Filter. The Peak Signal to Noise Ratio (PSNR) value for each algorithm has been compared with this proposed algorithm. A comparative analysis has been carried out in this paper between these predefined methods and the proposed algorithm. The tests have been done on few images and the results have been displayed in tabular format.