基于改进Census变换的自由曲面立体匹配方法

doi:10.3969/j.issn.1006-2475.2015.07.010

计算机与现代化

基于改进Census变换的自由曲面立体匹配方法

1.江南大学机械工程学院，江苏无锡214122； 2.江苏省食品先进制造装备技术重点实验室，江苏无锡214122；
3.江南大学物联网工程学院，江苏无锡214122

收稿日期:2015-02-11 出版日期:2015-07-23 发布日期:2015-07-28
作者简介:化春键(1975-)，男，北京人，江南大学机械工程学院副教授，博士，研究方向：机器视觉与传感器技术；方程骏(1990-)，男，浙江杭州人，硕士研究生，研究方向：机器视觉。
基金资助:
国家自然科学基金资助项目(61104213)；中央高校基本科研业务费专项资金资助项目(JUSRP11008)

A Freeform Surface Stereo Matching Method Based on Improved Census Transform

1. School of Mechanical Engineering, Jiangnan University, Wuxi 214122, China；

2. Jiangsu Key Laboratory of Advanced Food Manufacturing Equipment & Technology, Wuxi 214122, China；
3. School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China

Received:2015-02-11 Online:2015-07-23 Published:2015-07-28

摘要/Abstract

摘要：

针对基于网格投影的自由曲面三维重建方法中，特征点集中分布于投影网格线条之上，而投影网格线条未能覆盖的部分即投影网格线条间的自由曲面区域无特征点，导致重建精度较低的问题，提

出一种基于自适应窗口的改进Census变换立体匹配方法，对投影网格线条间的自由曲面进行三维重建；将特征匹配算法与区域匹配算法相结合，以图像中目标物体上的投影网格指导Census变换；利用可

变视差范围计算Hamming距离作为匹配代价；通过筛选获得精度较高的匹配点对。实验结果表明，该算法应用于双目视觉图像三维重建，可有效提高投影网格线条间自由曲面的重建精度。

关键词: 网格投影, Census变换, 自由曲面, 三维重建, 双目视觉

Abstract:

The feature points obtained from freeform surface 3D reconstruction method which based on grid projection all distribute along the projected grid lines only, and

there is none feature point on the surface among the projected grid lines, which lead to the low accuracy reconstruction. An improved Census transform stereo matching method

based on adaptive window is presented to reconstruct the surface among the projected grid lines. The method incorporates featurebased method and regionbased method. The grid

lines projected on the experimental subject surface are utilized to guide Census transform, the Hamming distance computed by dynamic disparity is used as the matching cost, and

the higher accuracy matching points are obtained after filtering. The proposed method is applied on the binary vision 3D reconstruction to validate the improved reconstruction

accuracy of surface among the projected grid lines.

Key words: grid projection, Census transform, freeform surface, 3D reconstruction, binocular vision

中图分类号:

TP391

化春键1,2，方程骏1,2，陈莹3. 基于改进Census变换的自由曲面立体匹配方法[J]. 计算机与现代化, doi: 10.3969/j.issn.1006-2475.2015.07.010.

HUA Chunjian1,2， FANG Chengjun1,2， CHEN Ying3. A Freeform Surface Stereo Matching Method Based on Improved Census Transform[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2015.07.010.

参考文献

［1］
马颂德,张正友. 计算机视觉［M］. 北京:科学出版社, 1998:7274.
［2］曹焱. 基于双目立体视觉信息的三维重建方法研究［D］. 长春:吉林大学, 2013.
［3］邱兆文，张田文. 文物三维重建关键技术［J］. 电子学报, 2008,36(12):2423-2427.
［4］ Goshtasby A, Stockman G C, Page C V. Aregionbased approach to digital image registration with subpixel accuracy［J］. IEEE Transactions on Geoscience and Remote Sensing,

1986,24(3):390-399.
［5］ Scharstein D, Szeliski R. A taxonomy and evaluation of twoframe stereo correspondence algorithm［J］. International Journal of Computer Vision, 2002,47(1):7-42.
［6］ Fusiello A, Roberto V. Efficient stereo with multiple windowing［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 1997:858-863.
［7］ Zhang K. Crossbased local stereo matching using orthogonal integral images［J］. IEEE Transactions on Circuits and Systems for Video Technology, 2009,19(7):1073-1079.
［8］ Zabih R, Woodfill J. Nonparametric local transforms for computer visual correspondence［C］// European Conference on Computer Vision. 1994:151158.
［9］ Banks J. Bennamoun M, Corke P. Nonparametric techniques for fast and robust stereo matching［C］// Proceedings of IEEE Ten CONSpeech and Image Technologies for

Computing and Telecommunications. 1997:365-368.
［10］Cyganek B, Borgosz J. A comparative study of performance and implementation of some areabased stereo algorithms ［C］// Computer Analysis of Images and Patterns, 9th

International Conference, CAIP 2001. 2001:709-716.
［11］Hirschmulle H, Scharstein D. Evaluation of cost functions for stereo matching［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2007:1

-8.
［12］Gonzalez R C, Woods R E, Eddins S L. 数字图像处理［M］. Matlab版. 北京:电子工业出版社, 2005.
［13］雷磊，郑江滨，宋雪梅．基于改进Census变换的立体匹配算法［J］．计算机应用研究， 2013,30(10):3185-3188.
［14］赖小波，朱世强，马璇．一种对光照条件不敏感而快速的局部立体匹配［J］．机器人, 2011,33(3):292-298.
［15］Ioannis Pitas, Venetsanopoulos A N. Nonlinear Digital Filters: Principles and Applications［M］. Boston: Kluwer Academic Publisher, 1990.
［16］宋雷. 基于条纹投影和条纹反射的三维形貌测量［D］. 成都:电子科技大学, 2012.
［17］Blanz V, Vetter T. A morphable model for the synthesis of 3D face［C］// Computer Graphics Proceedings, Annual Conference Series, ACM SIGGRAPH. 1999:187-194.

[1]	张亚文, 林文忠, 韩晓东. 基于组合特征点和主成分分析的点云配准算法[J]. 计算机与现代化, 2023, 0(09): 59-63.
[2]	张文元, 海琳琦, 刘英杰, 张海波. 一种基于改进多光谱的稀疏角CB-XLCT成像方法[J]. 计算机与现代化, 2022, 0(05): 96-101.
[3]	苏桃, 邵盈鑫, 李纯斌, 吴静, 常秀红, 杨帆. 基于运动恢复结构方法的植株三维重建[J]. 计算机与现代化, 2020, 0(11): 94-99.
[4]	孟兆旭，蔡超. 环境重建三维格网数据下的快速航迹规划方法[J]. 计算机与现代化, 2017, 0(11): 17-22+34.
[5]	韦羽棉，尚赵伟. 基于Kinect的旋转刚体三维重建方法[J]. 计算机与现代化, 2014, 0(5): 89-93，98.
[6]	董小龙;郭太良. 一种基于激光散斑去相关技术的三维重建方法[J]. 计算机与现代化, 2012, 203(7): 83-87.
[7]	党光磊;韦巍. 基于MITK的医学图像系统设计[J]. 计算机与现代化, 2011, 1(2): 8-10.
[8]	沈文超;孙涌. 不规则扁平粒状物表面平整度识别与分选方法的研究 [J]. 计算机与现代化, 2010, 1(7): 115-119.

基于改进Census变换的自由曲面立体匹配方法

A Freeform Surface Stereo Matching Method Based on Improved Census Transform

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 8

编辑推荐

Metrics

本文评价