SIA OpenIR  > 机器人学研究室
基于双目视觉与切面向量的曲面物体位姿估计
Alternative TitlePose Estimation of Curved Objects Based on Binocular Vision and Vectors of the Tangent Plane
刘玉珍1; 张嘉蓉1; 林森1,2,3
Department机器人学研究室
Source Publication激光与光电子学进展
ISSN1006-4125
2019
Pages1-19
Contribution Rank2
Funding Organization国家自然科学基金(61473280,61773367);国家自然科学基金委员会“共融机器人基础理论与关键技术研究”重大研究计划(91648118);辽宁省教育厅科学研究一般项目(L2014132);辽宁省自然科学基金面上项目(2015020100)
Keyword机器视觉 双目视觉 位姿估计 曲面物体 合作靶标
Abstract针对单目视觉估计曲面物体位姿时世界坐标不易获得的问题,将双目视觉与合作靶标相结合,提出了一种高效的曲面物体位姿估计方法。通过双目相机生成不同位姿下的目标物体点云以便快速提取靶标角点的世界坐标,不同于常用的点云配准位姿估计,本文取对应点坐标差的均值表示平移向量;然后求取靶标角点所在切面的法向量,组成目标在不同位姿坐标系下的矩阵,从而推导出旋转矩阵。所提方法不仅保证了位姿估计结果的准确性与稳定性,而且显著提高了算法的运行效率,实验结果表明,本文方法的运行效率相比ICP算法提高了98.24%,比NDT算法提高了97.58%,具有实际应用价值。
Other AbstractAiming at the problem that the world coordinates are not easy to obtain when estimating the pose of curved objects by monocular vision, an efficient method for estimating the pose of curved objects is proposed, which combines binocular vision and the cooperative target. A point cloud of the target object in different pose is generated by a binocular camera to quickly extract the world coordinates of the target corner points. Different from the commonly used point cloud registration pose estimation, the mean value of the corresponding point coordinate difference is used to represent the translation vector; then the normal vector of the tangent plane composed of the target corner points is taken to form a matrix of the target in different pose coordinates, thereby the rotation matrix is derived. The proposed method not only ensures the accuracy and stability of the pose estimation results, but also significantly improves the efficiency of the algorithm. The experimental results show that the efficiency of the proposed method is 98.24% higher than that of the ICP algorithm and 97.58% higher than the NDT algorithm, the proposed method has practical application value.
Language中文
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/25456
Collection机器人学研究室
Corresponding Author张嘉蓉
Affiliation1.辽宁工程技术大学电子与信息工程学院
2.中国科学院沈阳自动化研究所机器人学国家重点实验室
3.中国科学院机器人与智能制造创新研究院
Recommended Citation
GB/T 7714
刘玉珍,张嘉蓉,林森. 基于双目视觉与切面向量的曲面物体位姿估计[J]. 激光与光电子学进展,2019:1-19.
APA 刘玉珍,张嘉蓉,&林森.(2019).基于双目视觉与切面向量的曲面物体位姿估计.激光与光电子学进展,1-19.
MLA 刘玉珍,et al."基于双目视觉与切面向量的曲面物体位姿估计".激光与光电子学进展 (2019):1-19.
Files in This Item: Download All
File Name/Size DocType Version Access License
基于双目视觉与切面向量的曲面物体位姿估计(1313KB)期刊论文出版稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[刘玉珍]'s Articles
[张嘉蓉]'s Articles
[林森]'s Articles
Baidu academic
Similar articles in Baidu academic
[刘玉珍]'s Articles
[张嘉蓉]'s Articles
[林森]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[刘玉珍]'s Articles
[张嘉蓉]'s Articles
[林森]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 基于双目视觉与切面向量的曲面物体位姿估计.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.