Decision Based Adaptive Gradient Mean Filter (DBAGM)

ABSTRACT The Objective of this study is to find a better method to remove Salt and Pepper noise from digital image. A new algorithm named Decision Based Adaptive Gradient Mean (DBAGM)is proposed in this study. It detects the salt and pepper noise from DN numbers of each pixel, calculates mean using noise free neighbour pixels and replace the DN number of the affected pixel. Based on performance evaluation using Peak Signal to Noise Ratio (PSNR) and Root Mean Square Error, the proposed method outperformsgeneral mean and median filter in Salt and Pepper noise removal. DBAGM algorithm is giving extraordinary performance even when the noise density is 70%. The important feature of DBAGM algorithm is that it doesnt account the noisy pixels while taking mean. Keywords: Salt and pepper noise, Peak Signal to Noise Ratio (PSNR),Root Mean Square Error (RMSE), Decision Based Adaptive Gradient Mean (DBAGM) and Digital Number (DN).