采用多级动态规划实现立体匹配

郭永彩; 魏朋玉; 高潮

doi:null

您当前的位置：

首页 >

文章列表页 >

采用多级动态规划实现立体匹配

信息科学 | 更新时间：2020-08-12

- 采用多级动态规划实现立体匹配
- Stereo correspondence using multi-stage dynamic programming
- 光学精密工程 2009年17卷第7期页码：1745-1751
- 作者机构：
  
  重庆大学光电技术及系统教育部重点实验室重庆,400030
- 作者简介：
- 基金信息：
  
  教育部重点科研项目(No.108174)()
- DOI：
  中图分类号： TP391.4
- 收稿日期：2008-08-11，
  
  修回日期：2008-09-01，
  
  网络出版日期：2009-07-25，
  
  纸质出版日期：2009-07-25
- 稿件说明：
移动端阅览
郭永彩, 魏朋玉, 高潮. 采用多级动态规划实现立体匹配[J]. 光学精密工程, 2009,17(7): 1745-1751

GUO Yong-cai, WEI Peng-yu, GAO Chao. Stereo correspondence using multi-stage dynamic programming[J]. Editorial Office of Optics and Precision Engineering, 2009,17(7): 1745-1751
郭永彩, 魏朋玉, 高潮. 采用多级动态规划实现立体匹配[J]. 光学精密工程, 2009,17(7): 1745-1751 DOI：

GUO Yong-cai, WEI Peng-yu, GAO Chao. Stereo correspondence using multi-stage dynamic programming[J]. Editorial Office of Optics and Precision Engineering, 2009,17(7): 1745-1751 DOI：

摘要

动态规划算法是立体匹配中比较流行的一种全局优化方法

然而传统的动态规划立体匹配算法因忽略了核线间像素对视差的影响

产生了比较明显的横向"条纹"效应。为了消除此效应

本文提出了一种采用多级动态规划的立体匹配算法。该算法通过建立初始视差空间

对核线间进行双向的动态规划运算

然后将合并得到的结果用于优化初始视差空间。基于新的视差空间在核线上进行双向的动态规划运算

最后求取使核线上和核线间动态规划合并结果最小的值作为所求视差。利用标准数据库Middlebury平台对提出的算法进行了测试。实验结果表明

该算法与传统的动态规划算法和基于扫描线优化算法相比

横向"条纹"效应有所改善

且总体误匹配率分别降低了28.60%和40.42%

提高了匹配结果的准确性。

Abstract

Dynamic programming is one of the oldest but still popular methods for stereo correspondence. Because the traditional stereo matching based on the dynamic programming has the well-known horizontal "streaking" effect for skiping the vertical edges

a stereo correspondence using the multi-stage dynamic programming is proposed to eliminate this effect in this paper. Firstly

the initial disparity space is constructed

and the down and up dynamic programming processes are operated across the epipolar lines. Then

the initial disparity space is optimized by the combined results above. Finally

based on the new disparity space

the forward and backward dynamic programming processes are implemented along the epipolar lines

and the disparities are obtained by minimizing the combined results of the dynamic programming along and across the epipolar lines. This algorithm is evaluated on the benchmark Middlebury database. The experimental results show that the "streaking" effect is suppressed and the general error matching rates have decreased by 28.60% and 40.42% respectively compared with that of the traditional dynamic programming algorithm and the scan-line optimization algorithm

which can offer a good trade off in terms of the accuracy. It is demonstrated that the proposed algorithm is more efficient.

关键词

Keywords

references

SCHARSTEIN D, SZWLISKI R. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms[J]. International Journal of Computer Vision, 2002,47(1-3):7-42.[2] 张春森. 三维运动分析中的运动-立体双匹配约束[J]. 光学精密工程,2007,15(6):945-950. ZHANG CH S. Motion-stereo double matching restriction in 3D movement analysis[J]. Opt. Precision Eng.,2007,15 (6):945-950. (in Chinese)[3] KIM J C, LEE K M, CHOI B T, et al.. A dense stereo matching using two-pass dynamic programming with generalized ground control points . Proceedings of IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, San Diego,CA,USA: IEEE Computer Society, 2005,2:1075-1082.[4] 邓毅,林学訚. 一种新的快速立体视觉导航算法[J]. 电子学报,2006,34(11):2090-2093. DENG Y, LIN X Y. A novel fast stereo algorithm for vision navigation[J].Acta Electronica Sinica,2006,34(11):2090-2093. (in Chinese)[5] 谢少荣,王东红,罗均,等. 基于生物信息学中双DNA序列比对算法的图像立体匹配及其实现[J]. 光学精密工程,2007,15(1):106-111. XIE SH R, WANG D H, LUO J, et al.. Novel stereo matching algorithm based on pair-wise DNA alignment algorithm in bioinformatics and its implementation[J]. Opt. Precision Eng. , 2007,15(1):106-111. (in Chinese)[6] VEKSLER O. Stereo correspondence by dynamic programming on a tree . Proceedings of IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, San Diego,CA,USA: IEEE Computer Society, 2005:384-390. [7] 姜凯,陈海霞,刘立峰,等. 基于模板抽样的快速图像匹配算法[J]. 光学精密工程,2004,12(3):311-315. JIANG K,CHEN H X,LIU L F, et al.. Fast image matching algorithm based on template sampling[J]. Opt. Precision Eng.,2004,12(3):311-315. (in Chinese)[8] INTILLE S S, BOLICK A F. Disparity-space images and large occlusion stereo . Proceedings of the Third European Conference on Computer Vision, Stockholm, Sweden: Springer Berlin/Heidelberg, 1994,2:179-186.[9] 尹传历,向长波,宋建中,等. 一种基于自适应窗口和图切割的快速立体匹配算法[J]. 光学精密工程,2008,16(6):1117-1121. YIN CH L, XIANG CH B, SONG J ZH, et al.. Fast stereo matching algorithm based on adaptive window and graph cuts[J]. Opt. Precision Eng., 2008,16(6):1117-1121.(in Chinese)[10] BIRCHFIELD S, TOMASI C. A pixel dissimilarity measure that is insensitive to image sampling[J]. IEEE Trans. on Pattern Analysis and Machine intelligence, 1998,20(4):401-406.

浏览量

456

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于种子点传播的快速立体匹配

LightDiffu-DCE：基于光照强度扩散的低光照图像增强

多级图特征融合引导相机位姿回归

基于机载的红外动态目标视频实时超分辨率重建

基于知识蒸馏的Transformer视觉跟踪器