PROJECT TITLE :

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

ABSTRACT:

In this paper, a completely unique technique to hurry-up a non-local suggests that (NLM) filter is proposed. In the first NLM filter, most of its computational time is spent on finding distances for all the patches within the search window. Here, we tend to build a dictionary in which patches with similar photometric structures are clustered along. Dictionary is built only 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 terribly quickly for economical NLM denoising. We have a tendency to achieve a substantial reduction in computational price compared with the first NLM technique, especially when the search window of NLM is large, without a lot of affecting the PSNR. Second, we have a tendency to show that by building a dictionary for edge patches versus intensity patches, it's doable to scale back the dictionary size; so, additional improving the computational speed and memory demand. The proposed technique preclassifies similar patches with the same distance live as used by NLM technique. The proposed algorithm is shown to outperform different prefiltering based mostly quick NLM algorithms computationally and qualitatively.