A New Improved Statistical Algorithm for Image Noise reduction

Abstract

Many filtering algorithms have good noise reduction methods regardless the time complexity. This paper proposed a statistical algorithm which use correlation of the image to develop the filtering function. The correlating function applies filtering mask over the image to reduce the degradation. It can calculative resize the mask according to noise levels of the mask. The statistical histogram method is also introduced in the searching process of the median value. Experimental results show that the algorithm reduces the noise a much as possible and retains the detailed pixel value of the image. The complexity of the algorithm is decreased and efficiency of filtering function has been improved.