基于凸集论的稠密视差图优化算法

吉林大学学报(工学版) ›› 2013, Vol. 43 ›› Issue (增刊1): 204-208.

基于凸集论的稠密视差图优化算法

高凯, 陈贺新, 赵岩, 耿英楠, 黎昌明

吉林大学通信工程学院,长春 130012

收稿日期:2012-06-05 发布日期:2013-06-01
作者简介:高凯(1982-),男,博士研究生.研究方向:立体视频处理.E-mail:kai_gao@126.com
基金资助:
国家自然科学基金资助项目(61171078;60832002);吉林大学杰出青年基金项目(200905018).

Dense disparity map optimization algorithm based on convex set theory

GAO Kai, CHEN He-xin, ZHAO Yan, GENG Ying-nan, LI Chang-ming

College of Communication Engineering, Jilin University, Changchun 130012, China

Received:2012-06-05 Published:2013-06-01

摘要/Abstract

摘要：

针对计算机视觉领域中立体匹配算法在图像无纹理区产生的误匹配问题,提出以全变差函数为凸约束集,使用次梯度投影法,对通过初始视差变换后的立体图像对之间的像素误差函数和视差二次平滑因子形成的严格凸函数进行求解,寻求立体图像对之间的最优视差图。实验结果表明,该方法不仅能保留初始视差图的边缘,而且减少了由先前立体匹配算法求得的视差图在无纹理区产生的误匹配点,取得了良好的视差图优化效果。

关键词: 立体匹配, 凸优化, 次梯度投影, 全变差, 二次平滑因子

Abstract:

According to the mismatch problem in less textured area of stereo match algorithms in computer vision field,in order to find the optimal disparity map of the stereo image pair,a method using the sub-gradient projection method to solve strict convex function which composed by the pixel error function of the shaped image pairs and quadratic smoothing factor of disparity map was proposed,which was restrained by the convex constraint set of total variation function.Experimental results show that the method can keep the edge of primal disparity map well,reduce the mismatch in the less textured area of disparity generated by primal stereo matching algorithm,and get the better optimization disparity map effectively.

Key words: stereo matching, convex optimization, sub-gradient projection, total variation, quadratic smoothing factor

中图分类号:

TN919.8

高凯, 陈贺新, 赵岩, 耿英楠, 黎昌明. 基于凸集论的稠密视差图优化算法[J]. 吉林大学学报(工学版), 2013, 43(增刊1): 204-208.

GAO Kai, CHEN He-xin, ZHAO Yan, GENG Ying-nan, LI Chang-ming. Dense disparity map optimization algorithm based on convex set theory[J]. 吉林大学学报(工学版), 2013, 43(增刊1): 204-208.

参考文献

[1] 葛亮,朱庆生,傅思思,等.改进的立体像对稠密匹配算法[J].吉林大学学报:工学版,2010,40(1):212-217.GE Liang, ZHU Qing-sheng, FU Si-si,et al.Improved image dense stereo matching algorithm[J].Journal of Jilin University (Engineering and Technology Edition),2010,40 (1):212-217.

[2] Kolmogorov K,Zabih R.Computing visual correspondence with occlusions using graph cut[C]//In Proc ICCV,2001:508-515.

[3] Klaus A,Sormann M,Karner K.Segment based stereo matching using belief propagation and self-adapting dissimilarity measure[C]//ICPR,2006:15-18.

[4] Lei C,Selzer J,Yang Y H.Region tree based stereo using dynamic programming optimiza- tion[C]//In Proc CVPR,2006:2378-2385.

[5] Thomas Pock,Thomas Schoenemann,Gottfried Graber,et al.A con- vex formulation of continuous multi-label problems[C]//ECCV 2008,Part III,LNCS 5304,2008:792-805.

[6] Slesareva N,Bruhn A,Weickert J.Optical flow goes stereo:A variational method for estimating discontinuity preserving dense disparity maps[C]//Proc 27th DAGM Symposium,2005:33-40.

[7] Wided Miled,Jean Christophe Pesquet,Michel Parent.Disparity map estimation using a total variation bound[C]//Proceeding of the 3rd Canadian Conference on Computer and Robot Vision,Quebec,Canada,2006:48-55.

[8] Combettes P L.Convex set theoretic image recovery by extrapolated iterations of parallel subgradient projection[J].IEEE Transactons on Image Processing,1997,6(4):493- 506.

[9] Combettes P L.A block iterative surrogate constraint splitting method for quadratic signal recovery[J].IEEE Trans Signal Proc,1997,51(7):1771- 1782.

[10] Scharstein D,Szeliski R.A taxonomy and evaluation of dense two-frame stereo correspondence algorithms[C]//In Proceedings of IEEE Workshop on Stereo and Multi-Baseline Vision,2002:131-140.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed