曲阜师范大学学报（自然科学版）Issue(1)：72-77,6.DOI:10.3969/j.issn.1001-5337.2016.1.072
最小二乘模型下图像恢复的原-对偶算法的研究
The Primal-dual Algorithm for Image Restoration Based on Least Square Model
何姣姣 1庹谦 1陈剑鸣1
作者信息
- 1. 昆明理工大学理学院，650500，云南省昆明市
- 折叠
摘要
Abstract
As the regularization method is used to carry out the process of image restoration,the regu-larization parameter is often needed.There should be different regularization parameter for the different im-age.The least square method is used to recover the fuzzy image,and it is not required to adjust the regulari-zation parameter,and the recovery process is simplified.In this paper,the first-order original-dual method is used to solve constraint least squares problem of image recovery,and an improved algorithm is put forward which using the method of overall substitution calculation.The experimental results show that the primal dual algorithm for least squares model using iterative computation time is greatly saved,and the improved primal-dual algorithm in the noise free image restoration effect is better than that of the primal dual algo-rithm.关键词
正则化参数/模糊图像/图像恢复/最小二乘算法/原-对偶算法Key words
regularization parameter/blur image/image restoration/the least square algorithm/primal-dual algorithm分类
信息技术与安全科学引用本文复制引用
何姣姣,庹谦,陈剑鸣..最小二乘模型下图像恢复的原-对偶算法的研究[J].曲阜师范大学学报（自然科学版）,2016,(1):72-77,6.
