Non-Local Image De-noising and Post Processing Using KL Transform

In this paper an efforts has been made to estimate pixel intensity based on information present in the whole image and thereby exploiting the presence of similar patterns and features. The method is called non-local image de-noising method. This estimates noise-free pixel intensity as a weighted average of all pixel intensities in the image, and the weights are proportional to the similarity between the local neighborhood of the pixel being processed and local neighborhoods of surrounding pixels. This non-local method works on the assumption that image contains an extensive amount of redundancy. But we can find many pixels which have very less number of pixels with similar neighborhood in an image, especially at edges and when the image is corrupted with higher noise variance. Because of this, de-noised image visual quality is less. For further improvement in visual quality of image at the edges KL transform based local filter is designed and applied.

Keywords— KL Transform; de-noising; non-local method; visual quality.