Non Local Means Image Denoising for Color Images Using PCA

Abstract

The goal of image denoising is to remove unwanted noise from an image. There are various methods for image denoising. The proposed algorithm is a variation of the nonlocal means (NLM) image denoising algorithm that uses principal component analysis (PCA) to achieve a higher accuracy while reducing computational load. Image neighborhood vectors are first projected onto a lower dimensional subspace using PCA. For color images RGB image neighborhood vectors are formed by concatenating image neighborhoods in the three color channels into a single vector. The dimensionality of this subspace is chosen automatically using parallel analysis. Consequently, neighborhood similarity weights for denoising are computed using distances in this subspace rather than the full space. The accuracy of NLM and the proposed algorithm are examined with respect to the choice of image neighborhood and search window sizes. Finally, we present a quantitative and qualitative comparison of the proposed algorithm versus NLM image denoising algorithm.