PROJECT TITLE :

Novel Speed Up Strategies for Non-Local Means Denoising With Patch and Edge Patch Based Dictionaries - 2014

ABSTRACT:

During this paper, a completely unique technique to hurry-up a non-local means (NLM) filter is proposed. In the original NLM filter, most of its computational time is spent on finding distances for all the patches in the search window. Here, we build a dictionary in that patches with similar photometric structures are clustered together. Dictionary is built solely once with high resolution images belonging to different scenes. Since the dictionary is well organized in terms of indexing its entries, it's used to go looking similar patches very quickly for economical NLM denoising. We achieve a substantial reduction in computational value compared with the original NLM methodology, especially when the search window of NLM is large, while not abundant affecting the PSNR. Second, we have a tendency to show that by building a dictionary for edge patches as opposed to intensity patches, it is attainable to reduce the dictionary size; therefore, additional improving the computational speed and memory requirement. The proposed method preclassifies similar patches with the same distance measure as used by NLM methodology. The proposed algorithm is shown to outperform other prefiltering primarily based quick NLM algorithms computationally and qualitatively.