Journal of Software, Vol 6, No 9 (2011), 1746-1754, Sep 2011
doi:10.4304/jsw.6.9.1746-1754

A Line Segments Matching Method based on Epipolar-line Constraint and Line Segment Features

Jie-xian Zeng, Sheng Yu, Xiang Fu, Chao-xia Li

Abstract


Line segments are ordinary in industrial scene, accurate line segments matching is a key step for many applications, such as 3-D reconstruction. A matching method based on epipolar-line constraint and line segment features is proposed. Firstly, the points on line segments between image pairs are matched by epipolar-line constraint. Secondly, geometric descriptor and gray value descriptor are used to describe the line segment features, then the two descriptors are combined into a feature vector, and Euclidean distance between vectors is used to achieve fine match. Experiment results show that the proposed method is accurate and fast.


Keywords


epipolar-line constraint; line segment featrurs; line segments matching

References


J. S. Park, “Interactive 3D reconstruction from multiple images: a primitive-based approach,” Pattern Recognition, 2005, vol. 16, pp. 2558-2571.

K. P. Wang, Y. Sang and Q. F. Yu, “An Image Matching Approach Based on the Invariant Moment of Slope Angle Histogram,” Journal of Computer Aided Design & Computer Graphics, 2009, vol. 21, pp. 388-93.

G. J. Wen and R. S. Wang, “A global algorithm for straight line stereo matching based on feature grouping,” Journal of Software, 2001, vol. 11, pp. 1660-1667.

H. P. Sang, M. L. Kyoung and U. L. Sang, “A line feature matching technique based on an eigenvector approach,” Computer Vision and Image Understanding, 2000, vol. 77, pp. 263-283.
http://dx.doi.org/10.1006/cviu.2000.0808

J. Ji, Ang Haisong etal., “Fast extraction and matching of straight lines on sequential images,” Journal of Nanjing University of Aeronautics & Astronautics, 2005, vol. 37, pp. 227-231.

M.I.A. Lourakis, S.T. Halkidis, S.C. Orphanoudakis, “Matching disparate views of planar surfaces using projective invariants,” Image Vision Comput. 2000, vol.18, pp. 673–683.
http://dx.doi.org/10.1016/S0262-8856(99)00071-2

B. Herbert, F. Vittorio, V.G. Luc, “Wide-baseline stereo matching with line segments,” IEEE International Conference on Computer Vision and Pattern Recognition, 2005, pp. 329-336.

N. J. Chen and J. G. Li, “Match straight line based on line support region,” Infrared and Laser Engineering, 2000, vol. 29, pp. 10-14.

T. Li and X. L. Liu, “A robust approach for extracting and matching straight line,” Computer Simulation, 2008, vol. 25, pp. 171- 173.

Y. Deng, X.Y. Lin, “A fast line segment based dense stereo algorithm using tree dynamic programming,” European Conference on Computer Vision, 2006, pp. 201-212.

Z. H. Wang and F. C. Wu, “Mean Standard Deviation Descriptor and Line Matching,” PR&AI, 2009, vol. 22, pp. 32-39.

J. Cha and R. H. Cofer, “Extended hough Transform for linear feature detection,” Pattern Recognition, 2006 vol. 39, pp. 1034-1043.
http://dx.doi.org/10.1016/j.patcog.2005.05.014

C. Baillard, C. Schmid and A. Zisserman, “Automatic line matching and 3D reconstruction of buildings from multiple views,” In ISPRS Conference on Automatic Extraction of GIS Objects from Digital Imagery, UK, 1999, pp. 69-80.


Full Text: PDF


Journal of Software (JSW, ISSN 1796-217X)

Copyright @ 2006-2012 by ACADEMY PUBLISHER – All rights reserved.