ABSTRACT

Many real time images facing a problem of salt and pepper noise contaminated,due to poor illumination and environmental factors. Many filters and algorithms are used to remove salt and pepper noise from the image, but it also removes image information. This paper proposes a new effective algorithm for diagnosing and removing salt and pepper noise is presented. The existing standard algorithms like Median Filter (MF), Weighted Median Filter (WMF), Standard Median Filter (SMF) and so on, will yield poor performance particularly at high noise density. The suggested algorithm is compared with the above said standard algorithms using the metrics Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) value.The proposed algorithm exhibits more competitive performance results at all noise densities. The joint sorting and diagonal averaging algorithm has lower computational time,better quantitative results and improved qualitative result by a better visual appearance at all noise densities.