Please wait a minute...
浙江大学学报(工学版)
J4  2011, Vol. 45 Issue (2): 247-252    DOI: 10.3785/j.issn.1008-973X.2011.02.009
计算机技术     
基于全局优化的图像修复及其在GPU上实现
刘建明1,2, 鲁东明1, 葛蓉1
1.浙江大学 计算机科学与技术学院,浙江 杭州 310027;2.江西师范大学 计算机信息工程学院,江西 南昌 330022
Global optimization based image inpainting and
its implementation on GPU
LIU Jian-ming1,2, LU Dong-ming1, GE Rong1
1.College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China; 2. Department of
Computer and Information Engineering, Jiangxi Normal University, Nanchang 330022,China
 全文: PDF  HTML
摘要：

针对目前贪婪修复算法可能存在修复效果视觉不一致以及全局优化修复算法中未考虑结构信息的情况,提出一种新的基于全局优化的快速图像修复算法.通过定义出改进的能量函数,把图像修复问题转化为全局优化问题,并采用基于非局部均值的状态标签裁减算法,大幅度减少图中每个节点可能的状态标签数,从而大幅度降低优化算法复杂度；同时,利用图形处理器（GPU）进行加速,进一步提高了运算速度.与其他贪婪合成和最优化修复方法相比,该方法速度更快且较好地保持了纹理和结构的整体一致性.
Abstract:

Since greedy image inpainting algorithms might cause visual inconsistency, and global optimization based inpainting algorithms didn’t consider structure information, a new fast image inpainting algorithm based on global optimization is proposed. The image inpainting was formulated as a global optimization problem by defining an improved energy function, and a non-local mean based label pruning algorithm was adopted to cut down the number of labels for each node so as to reduce the complexity of the optimization algorithm significantly. Graphic processing unit (GPU) was used to further improve the computing speed. Compared with greedy synthesis and global optimization based image inpainting methods, the proposed approach not only avoids the inter-block discontinuity but also achieves better efficiency.
出版日期: 2011-03-17
:  TP 391.41  
基金资助:

长江学者和创新团队发展计划资助项目（IRT0652）；新世纪优秀人才支持计划资助项目(NCET-04-0535)；古代壁画保护国家文物局重点科研基地开放课题资助项目.
通讯作者: 鲁东明，男，教授, 博导.     E-mail: ldm@zju.edu.cn
作者简介: 刘建明（1981—），男，江西鹰潭人，博士，从事图像处理与识别研究.E-mail: liujianming@zju.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章　　

引用本文:

刘建明, 鲁东明, 葛蓉. 基于全局优化的图像修复及其在GPU上实现[J]. J4, 2011, 45(2): 247-252.

LIU Jian-ming, LU Dong-ming, GE Rong. Global optimization based image inpainting and
its implementation on GPU. J4, 2011, 45(2): 247-252.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2011.02.009    或    http://www.zjujournals.com/eng/CN/Y2011/V45/I2/247

［1］ BERTALMIO M, BERTOZZI A L, BALLESTER C, et al. Image inpainting ［C］∥Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques. New Orleans, Louisiana, USA: ACM, 2000: 417-424.
［2］ CHAN T F, SHEN J. Mathematical models for local nontexture inpaintings ［J］. SIAM Journal on Applied Mathematics. 2002, 62 (3): 1019-1043.
［3］ CHAN T F, SHEN J. Nontexture inpainting by curvaturedriven diffusions ［J］. Visual Communication and Image Representation, 2001, 12(4): 436-449.
［4］ ESEDOGLU S, SHEN J. Digital inpainting based on the mumford–shah–euler image model ［J］. European Journal of Applied Mathematics, 2002, 13(4): 353-370.
［5］ 侯正信,何宇清,许微.一种快速的图像修复算法［J］.中国图象图形学报.2007,21(10): 1909-1912.
HOU Zhengxin, HE Yuqing, XU Wei. A fast algorithm of image inpainting［J］.Journal of Image and Graphics,2007,21(10): 1909-1912
［6］ CRIMINISI A, PEREZ P, TOYAMA K. Region filling and object removal by exemplarbased image inpainting［J］. IEEE Transactions on Image Processing, 2004, 13(9): 1200-1212.
［7］ WU J, RUAN Q. Object removal by cross isophotes exemplarbased inpainting ［C］∥Proceedings of the 18th International Conference on Pattern Recognition. Hong Kong: IEEE Computer Society, 2006,3: 810-813.
［8］ GROVER S, MITTAL A, GUPTA S, et al. A unified approach for digital image inpainting using bounded search space ［J］. International Journal on Graphics, Vision and Image Processing, 5(6): 17-24.
［9］ SUN J, SHUM H Y. Image completion with structure propagation ［J］. ACM Transactions on Graphics, 2005, 24(3): 861 -868.
［10］ KOMODAKIS N, TZIRITAS G. Image completion using efficient belief propagation via priority scheduling and dynamic pruning ［J］. IEEE Transactions on Image Processing. 2007, 16(11): 2649-2661.
［11］ HUANG T, CHEN S, LIU J, et al. Image inpainting by global structure and texture propagation ［C］∥ Proceedings of the 15th international conference on Multimedia. Augsburg, Germany: ACM, 2007, 901: 517-520.
［12］ HAYS J, EFROS A A. Scene completion using millions of photographs ［J］. ACM Transactions on Graphics, 2007,26(3): 4-10
［13］ FELZENSZWALB P F, HUTTENLOCHER D P. Efficient belief propagation for early vision ［J］. International Journal of Computer Vision. 2006, 70(1): 41-54.
［14］ BUADES A, COLL B, MOREL J M. A nonlocal algorithm for image denoising ［C］∥Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, CA, USA: IEEE Computer Society, 2005, 2: 60-65.
［15］ NVIDIA C. Compute unified device architecture programming guide ［M］. Santa Clara, CA: Nvidia Coporation, 2007: 7-11.
[1] 杨玉婷, 史玉回, 夏顺仁. 基于讨论机制的头脑风暴优化算法[J]. J4, 2013, 47(10): 1705-1711.
[2] 朱晓恩, 郝欣, 夏顺仁. 基于Levy flight的特征选择算法[J]. J4, 2013, 47(4): 638-643.
[3] 孙创日,甄帅,夏顺仁. 基于吸引区域的多模态脑磁共振图像仿射配准[J]. J4, 2012, 46(9): 1722-1728.
[4] 谢迪, 童若锋, 唐敏, 冯阳. 具有高区分度的视频火焰检测方法[J]. J4, 2012, 46(4): 698-704.
[5] 李启雷, 金文光, 耿卫东. 基于无线惯性传感器的人体动作捕获方法[J]. J4, 2012, 46(2): 280-285.
[6] 戴渊明, 韦巍, 林亦宁. 基于颜色纹理特征的均值漂移目标跟踪算法[J]. J4, 2012, 46(2): 212-217.
[7] 刘晨彬,潘颖,张海石,黄峰平,夏顺仁. 基于磁共振图像的脑瘤MGMT表达状况检测算法[J]. J4, 2012, 46(1): 170-176.
[8] 钱诚, 张三元. 适用于目标跟踪的加权增量子空间学习算法[J]. J4, 2011, 45(12): 2240-2246.
[9] 曹颖, 郝欣, 朱晓恩, 夏顺仁. 基于自动随机游走的乳腺肿块分割算法[J]. J4, 2011, 45(10): 1753-1760.
[10] 吕谷来, 李建平, 李锵, 俞利兴, 朱松明, 楼建忠, 袁祎琳. 基于机器视觉的砧木定位识别方法[J]. J4, 2011, 45(10): 1766-1770.
[11] 赖小波,朱世强. 基于互相关信息的非参数变换立体匹配算法[J]. J4, 2011, 45(9): 1636-1642.
[12] 王金德, 寿黎但, 李晓燕, 陈刚. 基于多重分割捆绑特征的目标图像检索[J]. J4, 2011, 45(2): 259-266.
[13] 战江涛,刘强,柴春雷. 基于三维模型与Gabor小波的人脸特征点跟踪方法[J]. J4, 2011, 45(1): 30-36.
[14] 梁文锋,项志宇. 鲁棒的PTZ摄像机目标跟踪算法[J]. J4, 2011, 45(1): 59-63.
[15] 宋坤坡, 夏顺仁, 徐清. 考虑小波系数相关性的超声图像降噪算法[J]. J4, 2010, 44(11): 2203-2208.