计算机与数字工程2017，Vol.45Issue(4)：731-734,744,5.DOI:10.3969/j.issn.1672-9722.2017.04.029
一个基于卷积稀疏表示的图像重构算法
Image Reconstruction AlgorithmBased on Convolution Sparse Representation
摘要
Abstract
In the application of sparse representation in image reconstruction, the traditional method is to compute a set of overlapping image blocks independently.Using convolution sparse representation, the whole image is seen as a whole, the of sparse coding and by the filter dictionary and corresponding characteristic response of the convolution sum instead of the product of traditional dictionary matrix and coding coefficients for image decomposition.In this paper, based on the sparse representation model of convolution, an image reconstruction algorithm is proposed, the input image is approximated by the alternating direction multiplier method(ADMM), and the characteristic response coefficient is obtained.The experimental results show that the sparse performance of the convolution decomposition mechanism is better, and it is more suitable for image reconstruction.关键词
稀疏表示/卷积稀疏表示/ADMM/图像重构Key words
sparse representation/convolution sparse representation/ADMM/image reconstruction分类
信息技术与安全科学引用本文复制引用
陈小陶,许道云..一个基于卷积稀疏表示的图像重构算法[J].计算机与数字工程,2017,45(4):731-734,744,5.基金项目
国家自然科学基金(编号:61262006)资助. （编号:61262006）
