面向低纹理图像的快速立体匹配

张来刚; 魏仲慧; 何昕; 孙群

doi:10.3788/YJYXS20132803.0450

您当前的位置：

首页 >

文章列表页 >

面向低纹理图像的快速立体匹配

成像技术与图像处理 | 更新时间：2020-08-12

- 面向低纹理图像的快速立体匹配
- New Stereo Matching Method Based Edge Extraction
- 液晶与显示 2013年28卷第3期页码：450-458
- 作者机构：
  
  1. 中国科学院长春光学精密机械与物理研究所, 吉林长春 130033
  2. 中国科学院大学, 北京 100049
  3. 聊城大学汽车与交通工程学院,山东聊城,252059
- 作者简介：
- 基金信息：
  
  山东省自然科学基金(No.ZR2012CO026,No.ZR2011EL038)()
- DOI：10.3788/YJYXS20132803.0450
  中图分类号： TN911.73
- 收稿日期：2012-03-01，
  
  修回日期：2012-03-25，
  
  网络出版日期：2012-04-09，
  
  纸质出版日期：2013-06-15
- 稿件说明：
移动端阅览
张来刚, 魏仲慧, 何昕, 孙群. 面向低纹理图像的快速立体匹配[J]. 液晶与显示, 2013,28(3): 450-458

ZHANG Lai-gang, WEI Zhong-hui, HE Xin, SUN Qun. New Stereo Matching Method Based Edge Extraction[J]. , 2013,28(3): 450-458
张来刚, 魏仲慧, 何昕, 孙群. 面向低纹理图像的快速立体匹配[J]. 液晶与显示, 2013,28(3): 450-458 DOI： 10.3788/YJYXS20132803.0450.

ZHANG Lai-gang, WEI Zhong-hui, HE Xin, SUN Qun. New Stereo Matching Method Based Edge Extraction[J]. , 2013,28(3): 450-458 DOI： 10.3788/YJYXS20132803.0450.

摘要

立体匹配中

低纹理区域容易产生匹配多义性

导致失配

为了解决低纹理匹配问题

通常采用增大聚合窗口或全局优化算法(如:动态规划算法)

但是此类算法会导致边缘处的视差模糊不清

因此

文章提出了一种基于边缘图像的快速立体匹配算法。首先

对立体图像对进行边缘检测和Sobel滤波;然后

基于Sobel滤波后的图像

先后计算水平聚合代价和垂直聚合代价;最后

利用WTA(Winter-Take-All)优化算法得到最终视差图。实验中

对视差稠密度和准确度进行了定量分析

左右一致性检验平均达标率超过了88%

结果表明

本算法使用边缘提取取代图像分割作为代价聚合向导

很好地解决了低纹理区域立体匹配问题

同时大大提高了匹配效率

获得了准确可靠的视差图

达到了自主导航系统的要求。

Abstract

Computing disparity images for stereo pairs of low texture images is a challenging task because matching costs inside low texture areas of the stereo pairs are almost similar. This problem can not be solved straightforwardly by increasing the size of aggregation windows or by using global optimization methods

e.g. dynamic programming

because those approaches will smooth depth discontinued boundaries as well. This paper proposes a new method that is able to robustly perform stereo matching for low texture stereo images. First

edge detection and Sobel filtering are performed for the images; second

edge maps is used to guide the aggregation of pixel matching costs; Finally

disparity computation and left-right validation(88%) are used to get the disparity maps. Experimental results demonstrate that the proposed method utilizes the edge maps computed from the stereo pairs to guide the cost aggregation process in stereo matching

and can produce a larger number of and a better accuracy of reliable disparities for low texture stereo images than the moving average method.

关键词

Keywords

references

何友兵,李大海,李良玉,等.彩色立体图像匹配算法研究 [J]. 液晶与显示 ,2007,22(4):418-422.[2] Scharstein D, Szeliski R. A taxonomy and evaluation of dense two-frame stereo correspondence algorithms [J]. International J. Computer Vision, 2002, 47(1/2/3): 7-42.[3] Sach L T, Hamamoto K,Atsuta K, et al. A new coarseto-fine method for computing disparity images by sampling disparityspaces [J].Trans. on Electronics, Information and Systems, 2009, 129(1): 103-111.[4] W.vander Mark, Gavrila D M. Real-time dense stereo for intelligent vehicles [J].Trans. on Intelligent Transport Systems, 2006, 7(1):38-50.[5] Se S, Brady M. Ground plane estimation, error analysis and applications [J]. Trans. on Robotics and Autonomous Systems, 2002, 39(2):59-71.[6] Labayrade R, Aubert D, Tarel J P. Real time obstacle detection in stereo vision on non-flat road geometry through v-disparity representation [J].Proc. of Intelligent Vehicle Symposium, 2002, (2):646-651.[7] Hautiere N, Labayrade R, Perrollaz M, et al. Road scene analysis by stereovision: a robust and quasi-dense approach[C]// Proc of Control, Automation, Robotics and Vision, Singapore:IEEE,2006:1-6.[8] Chumerin Nikolay, Van Hulle M M. Ground plane estimation based on dense stereo disparity[C]// The Fifth International Conference on Neural Networks and Artificial Intelligence, Minsk, Belarus,2006:1-6.[9] Zanin M. Localization of ahead vehicle with on-board stereo cameras[C]//Proc. of Image Analysis and Processing, Modena:IEEE,2007:111-116.[10] Suganuma N, Fujiwara N. An obstacle extraction method using virtual disparity image[C]// Proc.of IEEE Intelligent Vehicles Symposium, Istanbl, Turkey:IEEE,2007:456-461.[11] Zhao Z, Katupitiya J, Ward J. Global correlation based ground plane estimation using v-disparity image[C]// Proc. of Robotics and Automation, Roma, Italy:IEEE, 2007: 529-534.[12] Kanade T, Okutomi M. Stereo matching algorithm with an adaptive window: theory and experiment [J]. Trans. on Pattern Analysis and Machine Intelligence, 1994, 16(9):920-932.[13] 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.[14] Gerrits M, Bekaert P. Local stereo matching with segmentation based outlier rejection[C]// Proc. of Computer and Robot Vision, IEEE,Canadian:IEEE, 2006:66-72.[15] Muhlmann K, Maier D, Hesser R, et al. Calculating dense disparity maps from color stereo images, an efficient implementation[C]// Proc. of Stereo and Multi-Baseline Vision, IEEE,Kauai:IEEE,2001:30-36.[16] Broggi A, Caraffi C, Fedriga R I, et al. Obstacle detection with stereo vision for off-road vehicle navigation[C]//Proc. of Computer Vision and Pattern Analysis, IEEE, San Diego,USA:IEEE,2005:65-72.[17] 吴晖辉,张宪民,洪始良.综合边缘和颜色特征的IC类贴装器件的检测 [J]. 光学精密工程, 2009,17(9):2283-2290.[18] Bank Jasmine, Corke Peter. Quantitative evaluation of matching methods and validity measures for stereo vision [J]. Trans. on International Journal of Robotics Research, 2001, 20(7):512-532.[19] 杨化超,张书毕,张秋昭. 基于SIFT的宽基线立体影像最小二乘匹配方法 [J]. 测绘学报, 2010,39(2):187-194.[20] Liang W, Mingwei G, Minglun G, et al. How far we can go with local optimization in real-time stereo matching-a performance study on different cost aggregation approaches[C]// Proc. of 3D Data Processing, Visualization, and Transmission, IEEE, ChapelHill:IEEE,2006:129-136.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于神经辐射场结构模糊优化的三维重建

基于全局优化的实时高精度模型重建

基于改进稳态匹配概率的立体匹配算法研究

基于块的背景重构算法研究及应用

基于改进MC算法的CT/MR图像三维重建