图像修复问题的低秩对偶逼近算法

doi:10.3969/j.issn.1001-2400.2016.04.030

西安电子科技大学学报 ›› 2016, Vol. 43 ›› Issue (4): 172-177.doi: 10.3969/j.issn.1001-2400.2016.04.030

图像修复问题的低秩对偶逼近算法

冯象初;王斯琪;李小平

(西安电子科技大学数学与统计学院，陕西西安 710071)

收稿日期:2015-05-04 出版日期:2016-08-20 发布日期:2016-10-12
通讯作者: 王斯琪(1987-)，女，西安电子科技大学博士研究生，E-mail：pearlwangxd@163.com．
作者简介:冯象初(1962-)，男，教授，E-mail：xcfeng@mail.xidian.edu.cn．
基金资助:
国家自然科学基金资助项目(61271294，61472303)；中央高校基本科研业务费专项资金资助项目(NSIY21)

Low-rank and dual approximation method for image inpainting problems

FENG Xiangchu;WANG Siqi;LI Xiaoping

(School of Mathematics and Statistics, Xidian Univ., Xi'an 710071, China)

Received:2015-05-04 Online:2016-08-20 Published:2016-10-12

摘要/Abstract

摘要：

为改进传统基于样本修复方法在实际应用中的不足，提出了一种新的图像修复算法．新算法以显著性排序法确保优先修复含明显结构边的目标块，利用图像欧氏距离搜索与该目标块匹配的相似样本块，对由搜索样本向量化构成的相似块矩阵进一步采用低秩对偶逼近提取可用信息以修复缺失像素．实验表明，新算法能够准确地优先修复显著性结构，且对多种类型的缺失均具有较好的修复效果．

关键词: 图像修复, 边界检测, 图像欧氏距离, 逼近理论

Abstract:

A novel image inpainting algorithm is proposed to improve the traditional exemplar-based inpainting methods. The new approach uses a salient-based ranking method to ensure the priority of the target patch with structure edges, whose similar exemplars would be measured by the image Euclidean distance. And a low-rank and dual approximation of the selected exemplar matrix is used to extract the available information for inpainting. Experimental results show that the new algorithm preferentially repairs significant structures accurately, and performs well in different kinds of missing or damaged cases.

Key words: image restoration, edge detection, image Euclidean distance, approximation theory

冯象初;王斯琪;李小平. 图像修复问题的低秩对偶逼近算法[J]. 西安电子科技大学学报, 2016, 43(4): 172-177.

FENG Xiangchu;WANG Siqi;LI Xiaoping. Low-rank and dual approximation method for image inpainting problems[J]. Journal of Xidian University, 2016, 43(4): 172-177.

参考文献

［1］ GUILLEMOT C, LE MEUR O. Image Inpainting: Overview and Recent Advances ［J］. IEEE Signal Processing Magazine, 2014, 31(1): 127-144.
［2］许建楼, 冯象初, 郝岩. 二阶总广义变分图像修复模型及其算法［J］. 西安电子科技大学学报, 2012, 39(5): 18-23.
XU Jianlou, FENG Xiangchu, HAO Yan. Second Order Total Generalized Variational Inpainting Model and Its Algorithm ［J］. Journal of Xidian University, 2012, 39(5): 18-23.
［3］吴玉莲, 冯象初. 利用平衡方法的非凸图像修复［J］. 西安电子科技大学学报, 2014, 41(5): 141-147.
WU Yulian, FENG Xiangchu. Nonconvex Image Inpainting via Balanced Regularization Approach ［J］. Journal of Xidian University, 2014, 41(5): 141-147.
［4］ CRIMINISI A, PREZ P, TOYAMA K. Region Filling and Object Removal by Exemplar-based Image Inpainting ［J］. IEEE Transactions on Image Processing, 2004, 13(9): 1200-1212.
［5］ BUYSSENS P, DAISY M, TSCHUMPERLE D, et al. Exemplar-based Inpainting: Technical Review and New Heuristics for Better Geometric Reconstructions ［J］. IEEE Transactions on Image Processing, 2015, 24(6): 1809-1824.
［6］ WOHLBERG B. Inpainting by Joint Optimization of Linear Combinations of Exemplars ［J］. IEEE Signal Processing Letters, 2011, 18(1): 75-78.
［7］ XU Z B, SUN J. Image Inpainting by Patch Propagation Using Patch Sparsity ［J］. IEEE Transactions on Image Processing, 2010, 19(5): 1153-1165.
［8］ DONG W S, SHI G M, LI X. Nonlocal Image Restoration with Bilateral Variance Estimation: a Low-rank Approach ［J］. IEEE Transactions on Image Processing, 2013, 22(2): 700-711.
［9］ WANG W W, ZHAI D, LI T, et al. Salient Edge and Region Aware Image Retargeting ［J］. Signal Processing: Image Communication, 2014, 29(10): 1223-1231.
［10］ WANG L W, ZHANG Y, FENG J F. On the Euclidean Distance of Images ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(8): 1334-1339.
［11］ WANG S Q, FENG X C. Optimization of the Regularization in Background and Foreground Modeling ［J］. Journal of Applied Mathematics, 2014: 592834.
［12］ GOLDSTEIN T, OSHER S. The Split Bregman Method for L1-regularized Problems ［J］. SIAM Journal on Imaging Sciences, 2009, 2(2): 323-343.
［13］ ROTH S, BLACK M J. Fields of Experts ［J］. International Journal of Computer Vision, 2009, 82(2): 205-229.
［14］ LI X. Variational Bayesian Image Processing on Stochastic Factor Graphs ［C］//Proceedings of 15th IEEE International Conference on Image Processing. Piscataway: IEEE, 2008: 1748-1751.

[1]	冯象初,王萍,何瑞强. 一种应用博弈和L0约束的盲图像修复方法[J]. 西安电子科技大学学报, 2021, 48(4): 103-112.
[2]	吴玉莲;冯象初. 利用平衡方法的非凸图像修复[J]. J4, 2014, 41(5): 141-147.
[3]	杨秀红;郭宝龙;严春满. 结构测度约束下基于加权分形的图像修复算法[J]. J4, 2012, 39(6): 84-91.
[4]	许建楼;冯象初;郝岩. 二阶总广义变分图像修复模型及其算法[J]. J4, 2012, 39(5): 18-23.
[5]	郝岩;冯象初;许建楼. 一种非局部扩散的图像修复模型[J]. J4, 2010, 37(5): 825-828.
[6]	田玉敏;赵缙;吴自力. 一种新的镜头边界检测算法 [J]. J4, 2008, 35(2): 288-292.
[7]	韩冰;姬红兵;高新波. 一种基于小波的分层和多分辨的镜头边界检测方法[J]. J4, 2005, 32(1): 36-40.

图像修复问题的低秩对偶逼近算法

Low-rank and dual approximation method for image inpainting problems

PDF (PC)

赞

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 7

Metrics

本文评价

推荐阅读 0