A novel pose estimation method based on three-view geometry constraint was proposed in order to solve camera pose estimation from image sequence. Image feature points were detected by using brute force selection mechanism, and feature points were matched based on optical flow algorithm and forward-backward error. Then pose estimation based on improved essential matrix decomposition was deduced. Object function was established according to epipolar geometry constraint, and rotation matrix was iteratively computed. Then a unique solution was obtained. The first and second camera projective matrices were obtained. The object function was built based on trifocal tensor and matching points, and the third camera pose matrix that was relative to the first camera was iteratively computed. Results show that the proposed method improves the robust, position accuracy and efficiency. The method can achieve a precision estimation fast and estimate the trajectory of rotor UAV.
Received: 07 September 2017
Published: 15 December 2017
ZHANG Zhen-jie, LI Jian-sheng, ZHAO Man-dan, ZHANG Xiao-dong. Camera pose estimation based on three-view geometry constraint. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2018, 52(1): 151-159.
[1] 陈明芽,项志宇,刘济林. 单目视觉自然路标辅助的移动机器人定位方法[J]. 浙江大学学报:工学版, 2014, 48(2):285-291. CHEN Ming-ya, XIANG Zhi-yu,LIU Ji-lin. Assistance localization method for mobile robot based on monocular natural visual landmarks[J]. Journal of Zhejiang University:Engineering Science, 2014, 48(2):285-291.
[2] AFIA A B,DEAMBROGIO L,SALOS D,et al. Review and classification of vision-based localization techniques in unknown environments[J]. IET Radar, Sonar and Navigation, 2014, 8(9):1059-1072.
[3] HEE L G, FAUNDORFER F, POLLEFEYS M. Motion estimation for self-driving cars with a generalized camera[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Portland:IEEE, 2013:2746-2753.
[4] BLOSCH M, WEISS S, SCARAMUZZA D, et al. Vision based MAV navigation in unknown and unstructured environments[C]//Robotics and automation (ICRA). Anchorage:IEEE, 2010:21-28.
[5] CHRISTAN F,MATIA P,DAVIDE S. SVO:fast semi-direct monocular visual odometry[C]//IEEE International Conference on Robotics and Automation (ICRA). Hong Kong:IEEE, 2014:15-22.
[6] ENGEL J, SCHOPS T, CREMERS D. LSD-SLAM:large-scale direct monocular SLAM[C]//European Conference on Computer Vision. Zurich:Springer, 2014:834-849.
[7] MUR-ARTAL R, MONTIEL J M M, TARDOS J D. Orb-slam:a versatile and accurate monocular slam system[J]. IEEE Transactions on Robotics, 2015, 31(5):1147-1163.
[8] HARTLEY R I, ZISSERMAN A. Multiple view geometry in computer vision[M]. UK:Cambridge University Press, 2004.
[9] 李国栋,田国会,王洪君,等. 弱标定立体图像对的欧式极线校正框架[J].光学精密工程, 2014, 22(7):1955-1961. LI Guo-dong, TIAN Guo-hui,WANG Hong-jun, et al. Euclidean epipolar rectification frame of weakly calibrated stereo pairs[J]. Optics and Precision Engineering, 2014, 22(7):1955-1961.
[10] 黄以君,刘伟军,赵吉宾. 一种基于三焦点张量的度量重建方法[J].仪器仪表学报,2009, 30(6):1307-1312. HUANG Yi-jun, LIU Wei-jun, ZHAO Ji-bin. Approach to metric reconstruction based on trifocal tensor[J]. Chinese Journal of Scientific Instrument, 2009, 30(6):1307-1312.
[11] GUERREIO J J, MUROLLO A C, SAGUES C. Localization and matching using the planar trifocal tensor with bearing-only data[J]. IEEE Transactions on Robotics, 2008, 24(2):494-501.
[12] LOPEZ-NICOLAS G, GUERRERO J J, SAGUES C. Visual control through the trifocal tensor for nonholonomic robots[J]. Robotics and Autonomous Systems, 2010, 58(2):216-226.
[13] KITT B, GEIGER A, LATEGAHN H. Visual odometry based on stereo image sequences with RANSAC-based outlier rejection scheme[C]//Intelligent Vehicles Symposium. La Jolla:IEEE, 2010:486-492.
[14] JIA B, CHEN J, ZHANG K. Adaptive visual trajectory tracking of nonholonomic mobile robots based on trifocal tensor[C]//2015 IEEE/RSJ International Conference on IEEE Intelligent Robots and Systems (IROS). Hamburg:IEEE, 2015:3695-3700.
[15] INDELMAN V, GURFIL P, RIBLIN E, et al. Real-time vision-aided localization and navigation based on three-view geometry[J]. IEEE Transactions on Aerospace and Electronic Systems, 2012, 48(3):2239-2259.
[16] KALAL Z, MIKOLAJCZYK K, MATAS J. Forward-backward error:automatic detection of tracking failures[C]//201020th International Conference on Pattern Recognition (ICPR). Istanbul:IEEE, 2010:2756-2759.
[17] 张跃强,苏昂,刘海波,等. 基于多直线对应和加权最小二乘的位姿估计[J].光学精密工程, 2015, 23(6):1722-1731. ZHANG Yue-qiang, SU Ang, LIU Hai-bo, et al. Pose estimation based on multiple line hypothesis and iteratively reweighted least squares[J]. Optics and Precision Engineering, 2015, 23(6):1722-1731.
[18] BAY H, TUYTELAARS T, VAN G L. Surf:speeded up robust features[C]//European Conference on Computer Vision. Graz:Springer, 2006:404-417.
WEI Xiao-feng, CHENG Cheng-qi, CHEN Bo, WANG Hai-yan. Chain code based on independent edge number[J]. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2018, 52(9): 1686-1693.
