PROJECT TITLE :

Image Denoising by Exploring External and Internal Correlations - 2015

ABSTRACT:

Single image denoising suffers from restricted knowledge collection at intervals a loud image. In this paper, we propose a completely unique image denoising theme, that explores each internal and external correlations with the assistance of net images. For every noisy patch, we tend to build internal and external knowledge cubes by finding similar patches from the noisy and web images, respectively. We tend to then propose reducing noise by a two-stage strategy using totally different filtering approaches. In the primary stage, since the noisy patch could lead to inaccurate patch selection, we have a tendency to propose a graph based optimization technique to improve patch matching accuracy in external denoising. The interior denoising is frequency truncation on internal cubes. By combining the internal and external denoising patches, we tend to acquire a preliminary denoising result. In the second stage, we tend to propose reducing noise by filtering of external and internal cubes, respectively, on remodel domain. In this stage, the preliminary denoising result not only enhances the patch matching accuracy but also provides reliable estimates of filtering parameters. The final denoising image is obtained by fusing the external and internal filtering results. Experimental results show that our technique constantly outperforms state-of-the-art denoising schemes in both subjective and objective quality measurements, e.g., it achieves >two dB gain compared with BM3D at a wide range of noise levels.