PROJECT TITLE :

Reweighted Low-Rank Matrix Analysis With Structural Smoothness for Image Denoising - 2018

ABSTRACT:

In this Project, we tend to develop a brand new low-rank matrix recovery algorithm for image denoising. We have a tendency to incorporate the full variation (TV) norm and therefore the pixel range constraint into the present reweighted low-rank matrix analysis to realize structural smoothness and to significantly improve quality in the recovered image. Our proposed mathematical formulation of the low-rank matrix recovery drawback combines the nuclear norm, TV norm, and l1 norm, thereby permitting us to use the low-rank property of natural pictures, enhance the structural smoothness, and detect and take away large sparse noise. Using the iterative alternating direction and quick gradient projection ways, we have a tendency to develop an algorithm to unravel the proposed difficult non-convex optimization downside. We conduct extensive performance evaluations on single-image denoising, hyper-spectral image denoising, and video background modeling from corrupted images. Our experimental results demonstrate that the proposed method outperforms the state-of-the-art low-rank matrix recovery methods, notably for massive random noise. For example, when the density of random sparse noise is thirty%, for single-image denoising, our proposed method is in a position to improve the standard of the restored image by up to four.twenty one dB over existing ways.