球面全景影像相对定向与精度验证

谢东海; 钟若飞; 吴俣; 符晗; 黄小川; 孙振兴

doi:10.11947/j.AGCS.2017.20160645

测绘学报 >

2017 , Vol. 46 >Issue 11: 1822 - 1829

DOI: https://doi.org/10.11947/j.AGCS.2017.20160645

摄影测量学与遥感

球面全景影像相对定向与精度验证

谢东海 ,
钟若飞 ,
吴俣 ,
符晗 ,
黄小川 ,
孙振兴

展开

1. 首都师范大学地球空间信息科学与技术国际化示范学院, 北京 100048;
2. 北京成像技术高精尖创新中心, 北京 100048;
3. 中国科学院遥感与数字地球研究所, 北京 100094

谢东海(1978-),男,博士,讲师,研究方向为摄影测量与遥感。E-mail:xdhbj@126.com

收稿日期: 2016-12-22

修回日期: 2017-08-14

网络出版日期: 2017-12-05

基金资助

国家自然科学基金（41371434）

收起

Relative Pose Estimation and Accuracy Verification of Spherical Panoramic Image

XIE Donghai ,
ZHONG Ruofei ,
WU Yu ,
FU Han ,
HUANG Xiaochuan ,
SUN Zhenxing

Expand

1. College of Geospatial Information Science and Technology, Capital Normal University, Beijing 100048, China;
2. Beijing Advanced Innovation Center for Imaging Technology, Beijing 100048, China;
3. Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100094, China

Received date: 2016-12-22

Revised date: 2017-08-14

Online published: 2017-12-05

Supported by

The National Natural Science Foundation of China (No. 41371434)

Fold

摘要

对传统的5点法相对定向算法的共面误差计算方法进行改进，提出了一种适合球面全景成像特点的相对定向计算流程。与传统方法相同，该算法首先计算本质矩阵，然后对本质矩阵进行奇异值分解得到旋转矩阵和平移矢量的候选解，最后利用重建的物方三维点坐标排除错误解。本文的贡献在于推导了球面全景共面条件公式，并使用点到核线平面的球面距离作为球面全景共面条件的误差项。模拟数据试验显示：当图像特征点的随机噪声在像素范围内时，3个姿态角的中误差约为0.1°，由相对定向恢复的相对平移量与模拟值的夹角中误差约为1.5°。使用车载全景相机配合POS获取的数据进行试验的结果显示：横滚角和俯仰角的中误差可以达到0.2°以内，航向角的中误差可以达到0.4°以内，由相对定向恢复的相对平移量与POS平移量的夹角中误差可以达到2°以内。采用本文相对定向算法的结果生成球面全景核线影像，提取影像之间同名点坐标并计算其列方向误差，结果显示核线影像同名点列坐标差的中误差在1个像素以内。

关键词： 球面全景影像; 相对定向; 核线; 本质矩阵; 奇异值分解

本文引用格式

谢东海 , 钟若飞 , 吴俣 , 符晗 , 黄小川 , 孙振兴 . 球面全景影像相对定向与精度验证[J]. 测绘学报, 2017 , 46(11) : 1822 -1829 . DOI: 10.11947/j.AGCS.2017.20160645

Abstract

This paper improves the method of the traditional 5-point relative pose estimation algorithm, and proposes a relative pose estimation algorithm which is suitable for spherical panoramic images. The algorithm firstly computes the essential matrix, then decomposes the essential matrix to obtain the rotation matrix and the translation vector using SVD, and finally the reconstructed three-dimensional points are used to eliminate the error solution. The innovation of the algorithm lies the derivation of panorama epipolar formula and the use of the spherical distance from the point to the epipolar plane as the error term for the spherical panorama co-planarity function. The simulation experiment shows that when the random noise of the image feature points is within the range of pixel, the error of the three Euler angles is about 0.1°, and the error between the relative translational displacement and the simulated value is about 1.5°. The result of the experiment using the data obtained by the vehicle panorama camera and the POS shows that:the error of the roll angle and pitch angle can be within 0.2°, the error of the heading angle can be within 0.4°, and the error between the relative translational displacement and the POS can be within 2°. The result of our relative pose estimation algorithm is used to generate the spherical panoramic epipolar images, then we extract the key points between the spherical panoramic images and calculate the errors in the column direction. The result shows that the errors is less than 1 pixel.

Key words： spherical panoramic images; relative pose estimation; epipolar line; essential matrix; SVD

参考文献

[1] 曾凡洋, 钟若飞, 宋杨, 等. 车载全景影像核线匹配和空间前方交会[J]. 遥感学报, 2014, 18(6):1230-1236. ZENG Fanyang, ZHONG Ruofei, SONG Yang, et al. Vehicle Panoramic Image Matching Based on Epipolar Geometry and Space Forward Intersection[J]. Journal of Remote Sensing, 2014, 18(6):1230-1236.
[2] 季顺平, 史云. 多镜头组合型全景相机两种成像模型的定位精度比较[J]. 测绘学报, 2014, 43(12):1252-1258. DOI:10.13485/j.cnki.11-2089.2014.0169. JI Shunping, SHI Yun. Comparison of Two Sensor Models for Multi-camera Rig System in Measurements[J]. Acta Geodaetica et Cartographica Sinica, 2014, 43(12):1252-1258. DOI:10.13485/j.cnki.11-2089.2014.0169.
[3] 石丽梅, 赵红蕊, 李明海, 等. 车载移动测图系统外方位元素标定方法[J]. 测绘学报, 2015, 44(1):52-58. DOI:10.11947/j.AGCS.2015.20130203. SHI Limei, ZHAO Hongrui, LI Minghai, et al. Extrinsic Calibration for Vehicle-based Mobile Mapping System[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44(1):52-58. DOI:10.11947/j.AGCS.2015.20130203.
[4] 张正鹏, 江万寿, 张靖. 自适应运动结构特征的车载全景序列影像匹配方法[J]. 测绘学报, 2015, 44(10):1132-1141. DOI:10.11947/j.AGCS.2015.20140622. ZHANG Zhengpeng, JIANG Wanshou, ZHANG Jing. An Matching Method for Vehicle-borne Panoramic Image Sequence Based on Adaptive Structure from Motion Feature[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44(10):1132-1141.
[5] 季顺平, 史云. 车载全景相机的影像匹配和光束法平差[J]. 测绘学报, 2013, 42(1):94-100, 107. JI Shunping, SHI Yun. Image Matching and Bundle Adjustment Using Vehicle-based Panoramic Camera[J]. Acta Geodaetica et Cartographica Sinica, 2013, 42(1):94-100, 107.
[6] 张正鹏, 江万寿, 张靖. 光流特征聚类的车载全景序列影像匹配方法[J]. 测绘学报, 2014, 43(12):1266-1273. DOI:10.13485/j.cnki.11-2089.2014.0172. ZHANG Zhengpeng, JIANG Wanshou, ZHANG Jing. An Image Match Method Based on Optical Flow Feature Clustering for Vehicle-borne Panoramic Image Sequence[J]. Acta Geodaetica et Cartographica Sinica, 2014, 43(12):1266-1273. DOI:10.13485/j.cnki.11-2089.2014.0172.
[7] PAGANI A, GAVA C, CUI Yan, et al. Dense 3D Point Cloud Generation from Multiple High-resolution Spherical Images.[C]//Proceedings of the 12th International Conference on Virtual Reality, Archaeology and Cultural Heritage. Prato, Italy:Eurographics Association, 2011:17-24.
[8] PAGANI A, STRICKER D. Structure from Motion Using Full Spherical Panoramic Cameras[C]//Proceedings of 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops). Barcelona, Spain:IEEE, 2012:375-382.
[9] TONG Guofeng, LIU Ran, TAN Jiandong. 3D Information Retrieval in Mobile Robot Vision Based on Spherical Compound Eye[C]//Proceedings of 2011 IEEE International Conference on Robotics and Biomimetics. Karon Beach, Phuket, Thailand:IEEE, 2011:1895-1900.
[10] LUHMANN T, TECKLENBURG W. 3-D Object Reconstruction from Multiple-Station Panorama Imagery[C]//MAAS H G, SCHNEIDER D. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XXXIV, Part 5/W16. Dresden, Germany:ISPRS, 2004.
[11] FANGI G. Further Developments of the Spherical Photogrammetry for Cultural Heritage[C]//Proceedings of the 22nd CIPA Symposium. Kyoto, Japan:CIPA, 2009.
[12] WAHBEH W, NEBIKER S, FANGI G. Combining Public Domain and Professional Panoramic Imagery for the Accurate and Dense 3D Reconstruction of the Destroyed Bel Temple in Palmyra[C]//ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2016, Volume Ⅲ-5. Prague, Czech Republic:ISPRS, 2016:81-88.
[13] FANGI G. The Multi-image Spherical Panoramas as A Tool for Architectural Survey[C]//XXI International CIPA Symposium. Athens, Greece:ISPRS, 2012, 36:256-1840.
[14] TONG Guofeng, WU Zizhang, WENG Ninglong, et al. An Omni-directional vSLAM Based on Spherical Camera Model and 3D Modeling[C]//Proceedings of the 10th World Congress on Intelligent Control and Automation. Beijing, China:IEEE, 2012:4551-4556.
[15] PAGANI A, HENRIQUES J, STRICKER D. Sensors for Location-based Augmented Reality the Example of Galileo and Egnos[J]. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLI-B1. Prague, Czech Republic:ISPRS, 2016:1173-1177.
[16] FANGI G, NARDINOCCHI C. Photogrammetric Processing of Spherical Panoramas[J]. The Photogrammetric Record, 2013, 28(143):293-311.
[17] TRIGGS B, MCLAUCHLAN P F, HARTLEY R I, et al. Bundle Adjustment:A Modern Synthesis[C]//TRIGGS B, ZISSERMAN A, SZELISKI R. Vision Algorithms:Theory and Practice. Berlin:Springer, 1999:298-372.
[18] 王佩军, 徐亚明. 摄影测量学[M]. 2版. 武汉:武汉大学出版社, 2010. WANG Peijun, XU Yaming. Photogrammetry[M]. 2nd ed. Wuhan:Wuhan University Press, 2010.
[19] HARTLEY R, ZISSERMAN A. Multiple View Geometry in Computer Vision[M]. 2nd ed. Cambridge:Cambridge University Press, 2000.
[20] HARTLEY R I. In Defense of the Eight-point Algorithm[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1997, 19(6):580-593.
[21] LONGUET-HIGGINS H C. A Computer Algorithm for Reconstructing a Scene from Two Projections[J]. Nature, 1981, 293(5828):133-135.
[22] NISTER D. An Efficient Solution to the Five-point Relative Pose Problem[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(6):756-770.
[23] FISCHLER M A, BOLLES R C. Random Sample Consensus:A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography[J]. Communications of the ACM, 1981, 24(6):381-395.
[24] 邓非, 吴幼丝, 胡玉雷, 等. 球形全景影像位姿估计的改进EPnP算法[J]. 测绘学报, 2016, 45(6):677-684. DOI:10.11947/j.AGCS.2016.20150411. DENG Fei, WU Yousi, HU Yulei, et al. Position and Pose Estimation of Spherical Panoramic Image with Improved EPnP Algorithm[J]. Acta Geodaetica et Cartographica Sinica, 2016, 45(6):677-684. DOI:10.11947/j.AGCS.2016.20150411.
[25] SZELISKI R. Computer Vision:Algorithms and Applications[M]. New York:Springer-Verlag, 2010.
[26] LOWE D G. Distinctive Image Features from Scale-invariant Keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110.

Options

文章导航

模态框（Modal）标题

摘要

本文引用格式

Abstract

参考文献