Skip to Main Content
One of the most popular image denoising methods based on self-similarity is called nonlocal means (NLM). Though it can achieve remarkable performance, this method has a few shortcomings, e.g., the computationally expensive calculation of the similarity measure, and the lack of reliable candidates for some nonrepetitive patches. In this paper, we propose to improve NLM by integrating Gaussian blur, clustering, and rotationally invariant block matching (RIBM) into the NLM framework. Experimental results show that the proposed technique can perform denoising better than the original NLM both quantitatively and visually, especially when the noise level is high.