[1]席志红,李爽,曾继琴,等.一种改进的PnP问题求解算法研究[J].应用科技,2018,45(04):56-60.[doi:10.11991/yykj.201612019]
 XI Zhihong,LI Shuang,ZENG Jiqin,et al.An improved algorithm for solving PnP problem[J].Applied science and technology,2018,45(04):56-60.[doi:10.11991/yykj.201612019]
点击复制

一种改进的PnP问题求解算法研究(/HTML)
分享到:

《应用科技》[ISSN:1009-671X/CN:23-1191/U]

卷:
第45卷
期数:
2018年04期
页码:
56-60
栏目:
计算机技术与应用
出版日期:
2018-08-05

文章信息/Info

Title:
An improved algorithm for solving PnP problem
作者:
席志红1 李爽12 曾继琴1 刁硕1
1. 哈尔滨工程大学 信息与通信工程学院, 黑龙江 哈尔滨 150001;
2. 中国电子科技集团公司 第五十四研究所, 河北 石家庄 050081
Author(s):
XI Zhihong1 LI Shuang12 ZENG Jiqin1 DIAO Shuo1
1. College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, China;
2. China Electronics Technology Group Corporation No. 54 th Research Institute, Shijiazhuang 050081, China
关键词:
透视n点投影问题计算机视觉不确定性旋转角度平移向量相机位姿稳定性鲁棒性
Keywords:
perspective-n-point projection problemcomputer visionuncertaintyrotation angletranslation vectorcamera posestabilityrobustness
分类号:
TP391.4
DOI:
10.11991/yykj.201612019
文献标志码:
A
摘要:
透视n点投影问题是计算机视觉领域的一个经典问题,有着广泛的应用前景。针对PnP问题中存在的位姿计算结果精度不高、稳定性不好,易受噪声干扰等问题,提出一种改进的PnP问题求解算法,该算法考虑了参考点的不确定性因素对位姿计算结果的影响,将不确定性描述矩阵加入到位姿求解等式中。采用模拟场景和真实场景进行实验验证,模拟场景中旋转角度误差小于0.2,平移向量误差小于0.1%;真实场景中平移向量平均误差为0.16 m。实验结果表明,该算法进一步提高了位姿求解精度,且具有良好的稳定性、鲁棒性和抗噪能力。
Abstract:
The perspective-n-point projection is a classical problem in the field of computer vision that has broad application prospect. Aiming at the low accuracy, unstable stability and susceptibility to noise and other problems in pose calculations of PnP problems, this paper proposes an improved algorithm for solving PnP problem. The algorithm takes into account the influence of uncertainties of the reference point on pose and adds the uncertainty description matrix to the equation for solving pose. The algorithm was validated by the simulated and real scene experiments. The error of rotation angle was less than 0.2 degrees, and the error of translation vector was less than 0.1% in simulated experiments, and the mean error of translation vector was 0.16m in real scene experiments. The experimental results show that the algorithm further improves the accuracy of pose solution, and has good stability, robustness and anti-noise ability.

参考文献/References:

[1] 席志红, 李爽, 甘兴利. PnP算法在室内定位中的应用[J]. 无线电工程, 2017, 47(10):39-44.
[2] 赵霞, 袁家政, 刘宏哲. 基于视觉的目标定位技术的研究进展[J]. 计算机科学, 2016, 43(6):10-16, 43.
[3] MVLLER M, RASSWEILER M C, KLEIN J, et al. Mobile augmented reality for computer-assisted percutaneous nephrolithotomy[J]. International journal of computer assisted radiology and surgery, 2013, 8(4):663-675.
[4] LEPETIT V, MORENO-NOGUER F, FUA P. EPnP:an accurate O(n) solution to the PnP problem[J]. International journal of computer vision, 2009, 81(2):155-166.
[5] YANG Sen, WU Fuchao. Weighted linear methods for camera pose estimation[J]. Journal of software, 2011, 22(10):2476-2487.
[6] FÖRSTNER W. Minimal representations for uncertainty and estimation in projective spaces[C]//Proceedings of the 10th Asian Conference on Computer Vision. Queenstown, New Zealand, 2010:619-632.
[7] FERRAZ L, BINEFA X, MORENO-NOGUER F. Very fast solution to the PnP problem with algebraic outlier rejection[C]//IEEE Conference on Computer Vision and Pattern Recognition, Columbus, USA, 2014:501-508.
[8] LI Shiqi, XU Chi, XIE Ming. A robust O(n) solution to the perspective-n-point problem[J]. IEEE transactions on pattern analysis and machine intelligence, 2012, 34(7):1444-1450.
[9] 李书杰, 刘晓平. 摄像机位姿的高精度快速求解[J]. 中国图象图形学报, 2014, 19(1):20-27.
[10] TRIGGS B. Camera pose and calibration from 4 or 5 known 3D points[C]//Proceedings of the Seventh IEEE International Conference on Computer Vision. Kerkyra, Greece, 1999:278-284.
[11] WU Yihong, HU Zhanyi. PnP problem revisited[J]. Journal of mathematical imaging and vision, 2006, 24(1):131-141.
[12] LU C P, HAGER G D, MJOLSNESS E. Fast and globally convergent pose estimation from video images[J]. IEEE transactions on pattern analysis and machine intelligence, 2000, 22(6):610-622.
[13] HESCH J A, ROUMELIOTIS S I. A direct least-squares (DLS) method for PnP[C]//IEEE International Conference on Computer Vision. Barcelona, Spain, 2011:383-390.

相似文献/References:

[1]赵龙,郭宝,李铁军.基于边缘增强模板匹配的PTZ主动目标跟踪系统[J].应用科技,2014,41(04):1.
 ZHAO Long,GUO Bao,LI Tiejun.Active camera tracking system for PTZ camera based on EETM algorithm[J].Applied science and technology,2014,41(04):1.
[2]王信,汪友生.基于深度学习与传统机器学习的人脸表情识别综述[J].应用科技,2018,45(01):65.[doi:10.11991/yykj.201707008]
 WANG Xin,WANG Yousheng.Facial expression recognition based on deep learning and traditional machine learning[J].Applied science and technology,2018,45(04):65.[doi:10.11991/yykj.201707008]

备注/Memo

备注/Memo:
收稿日期:2016-12-30。
基金项目:“十三五”国家重点研发计划项目(SQ2016YFGX040104).
作者简介:席志红(1965-),女,教授.
通讯作者:李爽,E-mail:1162854787@qq.com
更新日期/Last Update: 2018-09-05