基于稀疏表示的图像去噪算法优化

摘要

图像去噪是图像复原中一个非常重要的环节，图像去噪是图像处理中最基本的问题，目的是为了取得视觉上的高质量图像。随着压缩感知的兴起与推广，越来越多的学者开始关注稀疏表示理论及其应用，基于稀疏表示的图像去噪成为近年来该领域的前沿研究课题。论文采用非局部集中稀疏表示图像复原模型，创新性的把改进的 K 均值算法和 PCA原理结合起来，形成学习字典。通过对比经典算法，发现这种算法能获得更高的峰值信噪比，提高了同质区域平滑性，并保留更多的纹理、边缘等细节特征，使得图像恢复获得更好的质量。%Image denoising is very important in image restoration ,it is aimed to get a high quality image on the visual . Image denoising is the most basic problem in image processing .With the rise and popularization of compressed sensing ,more and more scholars begin to pay attention to the sparse representation theory and its applications ,image denoising based on image sparse representation has become a frontier research topics in this field in recent years .In this paper ,the model of nonlocally focused sparse representation is adopted for image restoration ,and the improved K-means algorithm and principle of PCA are combined creatively to form the atomic learning dictionary .By comparison with the classical algorithm ,it is found K-means that the algorithm can get higher PSNR and improve the smoothness of homogeneous regions while preserving edge and texture ,more features and other details ,and thus obtain better quality of image restoration .