首页 > 论文 > 激光与光电子学进展 > 56卷 > 23期(pp:231504--1)

基于对极约束的双目立体视觉标定精度评价方法

Binocular Stereo Vision Calibration Accuracy Evaluation Using Epipolar Constraint

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

为解决现有利用左右图像总残差的均值来评价双目立体视觉标定不够精确的问题,提出一种基于对极约束的双目立体视觉标定精度评价方法。该方法充分考虑双目立体视觉中左右图像特征的约束关系及相机标定参数的全局性特征。遵循最小匹配代价原则,该方法利用尺度不变特征变换立体特征匹配法来进行角点检测和匹配。通过左右图像平面上实测角点与其在相对图像平面上对应极线的匹配程度来评价双目立体视觉标定精度,并将这种算法加入到标定算法中,实现了在标定实验过程中对相机标定精度的实时评价。实验表明,该方法比总残差均值法精度更高,精度最高提高了54.0%。

Abstract

The accuracy of binocular stereo vision calibration obtained using the mean of total residuals of left and right images is unsatisfactory. To overcome this limitation, this study proposes a method for evaluating the accuracy of binocular stereo vision calibration based on the epipolar constraint. The proposed method considers the constraint relationship between left and right image features in binocular stereo vision and global characteristics of camera calibration parameters. Based on the principle of minimum matching cost, a stereo feature matching method based on scale-invariant feature transform is used for corner detection and matching. The accuracy of binocular stereo vision calibration is evaluated based on the matching degree of measured corner points on the left and right image planes with their corresponding epipolar lines on the relative image plane. This proposed algorithm is added to the calibration algorithm to realize real-time evaluation of camera calibration accuracy during the calibration experiment. Experiments demonstrate that the proposed method is more accurate than the method using the mean of residuals, with accuracy increased by up to 54.0%.

Newport宣传-MKS新实验室计划
补充资料

DOI:10.3788/LOP56.231504

所属栏目:机器视觉

基金项目:国家自然科学基金、陕西省自然科学基础研究计划、中央高校基本科研业务费专项基金;

收稿日期:2019-04-11

修改稿日期:2019-06-24

网络出版日期:2019-12-01

作者单位    点击查看

张青哲:长安大学道路施工技术与装备教育部重点实验室, 陕西 西安 710064
王勇:长安大学道路施工技术与装备教育部重点实验室, 陕西 西安 710064

联系人作者:张青哲(zqzh@chd.edu.cn)

备注:国家自然科学基金、陕西省自然科学基础研究计划、中央高校基本科研业务费专项基金;

【1】Jiang T, Cheng X S, Cui H H, et al. Calibration method of binocular vision system with zoom lens based on homography matrix [J]. Acta Optica Sinica. 2018, 38(3): 0315005.
姜涛, 程筱胜, 崔海华, 等. 基于单应性矩阵的变焦双目视觉标定方法 [J]. 光学学报. 2018, 38(3): 0315005.

【2】Zhu F, Yu F S, Wu Y M, et al. Analysis of attitude calibration precision of P4P camera [J]. Acta Optica Sinica. 2018, 38(11): 1115005.
朱帆, 于芳苏, 吴易明, 等. P4P法相机姿态标定精度分析 [J]. 光学学报. 2018, 38(11): 1115005.

【3】Zou P P, Zhang Z L, Wang P, et al. Binocular camera calibration based on collinear vector and plane homography [J]. Acta Optica Sinica. 2017, 37(11): 1115006.
邹朋朋, 张滋黎, 王平, 等. 基于共线向量与平面单应性的双目相机标定方法 [J]. 光学学报. 2017, 37(11): 1115006.

【4】Kawabata S, Kawai Y. Plane based multi camera calibration under unknown correspondence using ICP-like approach . [C]//Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012), November 11-15, 2012, Tsukuba, Japan. New York: IEEE. 2012, 3700-3703.

【5】Zhang G L, Yao E L, Lin Z L, et al. Fast binocular SLAM algorithm combining the direct method and the feature-based method [J]. Robot. 2017, 39(6): 879-888.
张国良, 姚二亮, 林志林, 等. 融合直接法与特征法的快速双目SLAM算法 [J]. 机器人. 2017, 39(6): 879-888.

【6】Zhang B S, Zhang Q Z, Wang Y S, et al. The method of solving the non-coplanar perspective-four-point (P4P) problem . [C]//Proceedings of the 33rd Chinese Control Conference, July 28-30, 2014, Nanjing, China. New York: IEEE. 2014, 1039-1043.

【7】Guo Y J, Bai F Z, Zhang T Y, et al. Ring object diameter measuring method based on Radon transform and gray gravity algorithm [J]. Laser & Optoelectronics Progress. 2015, 52(8): 081501.
郭延杰, 白福忠, 张铁英, 等. 基于Radon变换与灰度重心法的环形目标直径测量方法 [J]. 激光与光电子学进展. 2015, 52(8): 081501.

【8】He J M, Qiu J, Liu C. Fusing feature point density and edge information for scene depth estimation [J]. Laser & Optoelectronics Progress. 2017, 54(7): 071101.
何建梅, 邱钧, 刘畅. 融合特征点密度与边缘信息的场景深度估计 [J]. 激光与光电子学进展. 2017, 54(7): 071101.

【9】Wong K H, Yu Y K, Fung H Y, et al. Robust and efficient pose tracking using perspective-four-point algorithm and Kalman filter . [C]//2017 International Conference on Mechanical, System and Control Engineering (ICMSC), May 19-21, 2017, St.Petersburg, Russia. New York: IEEE. 2017, 240-244.

【10】Xie Z X, Lu W J, Wang X M, et al. Analysis of pose selection on binocular stereo calibration [J]. Chinese Journal of Lasers. 2015, 42(2): 0208003.
解则晓, 陆文娟, 王晓敏, 等. 双目立体标定的姿态选择分析 [J]. 中国激光. 2015, 42(2): 0208003.

【11】Tsai R. A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses [J]. IEEE Journal on Robotics and Automation. 1987, 3(4): 323-344.

【12】Qu H, Wu Z N. Step calibration and accuracy analysis of binocular stereo vision [J]. Journal of Tianjin Polytechnic University. 2018, 37(3): 67-71.
曲华, 吴朝娜. 双目立体视觉分步标定及精度分析 [J]. 天津工业大学学报. 2018, 37(3): 67-71.

【13】Weng J, Cohen P, Herniou M. Camera calibration with distortion models and accuracy evaluation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1992, 14(10): 965-980.

【14】Wu F Z. Mathematical methods in computer vision[M]. Beijing: Science Press, 2008, 63-67.
吴福朝. 计算机视觉中的数学方法[M]. 北京: 科学出版社, 2008, 63-67.

【15】Hartley R, Zisserman A. Multiple view geometry in computer vision[M]. UK: , 2004, 158-179.

引用该论文

Zhang Qingzhe,Wang Yong. Binocular Stereo Vision Calibration Accuracy Evaluation Using Epipolar Constraint[J]. Laser & Optoelectronics Progress, 2019, 56(23): 231504

张青哲,王勇. 基于对极约束的双目立体视觉标定精度评价方法[J]. 激光与光电子学进展, 2019, 56(23): 231504

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF