Image de-noising based on optimized NLM algorithm

Images and video are often coded using block based discrete cosine transform (DCT) or discrete wavelet transform (DWT) that cause a great deal of visual distortions. Non-Local Means (NLM) algorithm is chosen by means of comparing complexity and quality of different algorithms and is considered to be the better algorithm for artifacts reduction. Besides, implementation of this algorithm is computationally intensive. In this note, improvements to the non-local means introduced in [3-8] are presented and very effective performance optimization approach is presented. This approach is based on additional memory usage for caching pixels distance in the image. We present the underlying framework and experimental results for video that is processed by NLM with different parameters.