Performance of A Hybrid Salt and Pepper Noise Removal Technique

Abstract

Salt and Pepper Noise is one of the most disturbing corruption for any images. Many types of Image Processing techniques have been applied to remove this type of noise and restore the original image. This paper is based on a comparative study of three filters. Here we have taken Mean Filter, Median Filter and their combination, i.e., the Hybrid Filter. We have taken sample images, added Salt and Pepper noise in them, passed them through these three filters and studied the outcomes.We have presented the histograms and the data chart of correlation of the noisy image with that of the restored one through these three filters. We have seen the Mean filter performs the best all the time. We should however make the detection and filtering more robust and adaptive to draw any definite conclusion.