IMAGE AND VIDEO ENHANCEMENT USING SPARSE CODING, BELIEF PROPAGATION AND MATRIX COMPLETION

Abstract

Super resolution as an exciting application in image processing was studied widely in the literature. This dissertation presents new approaches to video super resolution, based on sparse coding and belief propagation. First, find candidate match pixels on multiple frames using sparse coding and belief propagation. Second, incorporate information from these candidate pixels with weights computed using the Nonlocal-Means (NLM) method in the first approach or using SCoBeP method in the second approach. The effectiveness of the proposed methods is demonstrated for both synthetic and real video sequences in the experiment section. In addition, the experimental results show that my models are naturally robust in handling super resolution on video sequences affected by scene motions and/or small camera motions. Moreover, in this dissertation, I describe a denoising method using low-rank matrix completion. In the proposed denoising approach, I present a patch-based video denoising algorithm by grouping similar patches and then formulating the problem of removing noise using a decomposition approach for low-rank matrix completion. Experiments show that the proposed approach robustly removes mixed noise such as impulsive noise, Poisson noise, and Gaussian noise from any natural noisy video. Moreover, my approach outperforms state-of-the-art denoising techniques such as VBM3D and 3DWTF in terms of both time and quality. My technique also achieves significant improvement over time against other matrix completion methods.