摘要
相较于传统综合字典学习方法,非参数贝叶斯方法具有显著优势,但其对图像结构全局相似性和变异性的表示能力仍有较大提升空间。针对这个问题,提出了一种基于结构相似性的非参数贝叶斯字典学习算法,该算法基于图像结构的全局相似性对图像进行聚类处理,并在图像的字典稀疏表示中引入块结构特性,提升了字典的结构表示能力。实验表明,所提算法在图像去噪和压缩感知方面的性能均优于目前主流的几种无监督字典学习算法。
Though nonparametric Bayesian methods possesses significant superiority with respect to traditional comprehensive dictionary learning methods, there is room for improvement of this method as it needs more consideration over the structural similarity and variability of images. To solve this problem, a nonparametric Bayesian dictionary learning algorithm based on structural similarity was proposed. The algorithm improved the structural representing ability of dictionaries by clustering images according to their non-local structural similarity and introducing block structure into sparse representing of images. Denoising and compressed sensing experiments showed that the proposed algorithm performs better than several current popular unsupervised dictionary learning algorithms.
作者
董道广
芮国胜
田文飚
康健
刘歌
DONG Daoguang;RUI Guosheng;TIAN Wenbiao;KANG Jian;LIU Ge(Signal and Information Processing Key Laboratory in Shandong, Navy Aviation University, Yantai 264001, China)
出处
《通信学报》
EI
CSCD
北大核心
2019年第1期43-50,共8页
Journal on Communications
基金
国家自然科学基金资助项目(No.41606117
No.41476089
No.61671016)~~
关键词
非参数贝叶斯
字典学习
结构相似性
图像去噪
压缩感知
nonparametric Bayesian
dictionary learning
structural similarity
denoising
compressed sensing