Abstract
3D reconstruction based on high-level features, such as line and plane, is a significant development trend in Digital Photogrammetry and Computer Vision. To auto-recognize corresponding line and point is a core problem in above research areas. Therefore this article presents a novel line segment stereo recognition algorithm based on straight line auto-grouping, epipolar line constraint guiding, vanishing point direction guiding, and image pyramid model. The strategy for auto-recognizing corresponding line bases on “independent observation” + “local integral constraint” + “large area search” + “probability relaxation”, so our algorithm satisfies objective law that the global distribution of space-object is discontinuous but local is continuous. Simultaneously, in order to overcome image geometric distortion, NCC similarity measure based on SIFT operator compensating is presented. Furthermore, in corresponding line recognition, the problems of discontinuties in space and variation of parallax can be solved. Finally, through comparing with method of homography line matching, experimental results prove that our algorithm has a better reliability, rightness and precision. Moreover this research paves the way for auto-3D-reconstruction.
Sponsored by the Key Program of Ministry of Education (No.108098) and the National Natural Science Foundation of China (No.40671078).
Civil Defense Industry Special Scientific Research Project (Objects / environmental electromagnetic sources of Theoretical and Key Technology Research).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Tien, F.-C., Sun, T.-H.: Solving Line-Feature Stereo Matching With Genetic Algorithms in Hough Space. Journal of the Chinese Institute of Industrial Engineers 21(5), 516–526 (2004)
Wang, Z.: Principles of Photogrammetry (with Remote Sensing), Press of Wuhan, Technical University of Surveying and Mapping, Publishing House of Surveying and Mapping (1990)
Zhang, Z., Zhang, J.: Digital Photogrammetry. Wuhan University Press, Wuhan (1997)
Leberl, F.W.: ISPRS Commission III “Towards Photogrammetric Computer Vision”. Highlights of ISPRS 6(2) (2001)
Zhang, Z.: Digital Photogrammetry and Computer Vision. Geomatics and Information Science of Wuhan University 29, 1035–1039 (2004)
Herbert, B., Vittorio, F., Luc, V.G.: Wide-Baseline Stereo Matching with LineSegments. In: IEEE International Conference on Computer Vision and Pattern Recognition (2005)
Sagues, C., Guerrero, J.: Robust line matching in image pairs of scenes with dominant planes. Optical Engineering (6), 067204/1–12 (2006)
Schmid, C., Zisserman, A.: The geometry and matching of lines and curves overmultiple views. International Journal of Computer Vision 40(3), 199–233 (2000)
Lourakis, M.I.A., Halkidis, S.T., Orphanoudakis, S.C.: Matching DisparateViews of Planar Surfaces Using Projective Invariants. Image and Vision Computing 18(9), 673–683 (2000)
Schmid, C., Zisserman, A.: Automatic line matching cross views. In: CVPR (1997)
Zhang, L., Zhang, Z., Zhang, J.: The Image Matching Based on Wallis Filtering. Journal of Wuhan Technical University of Surveying and Mapping (WTUSM) 24(1), 24–27 (1999)
Li, C., Wu, H., Hu, M., Zhou, Y.: A Novel Method of Straight-Line Extraction Based on Wallis Filtering for the Close-Range Building. In: The Asia Pacific Conference on Postgraduate Research in Microelectronics & Electronics (PrimeAsia), Shanghai, China, November 18-20 (2009)
Gruen, A.: Adaptive least square correlation: A powerful image matching technique. ISPRS Journal of Photogrammetry, Remote Sensing and Cartography 14(3), 175–187 (1985)
Zhang, Z.: From Digital Photogrammetry Workstation (DPW) to Digital Photogrammetry Grid (DPGrid). Geomatics and Information Science of Wuhan University 32, 565–571 (2007)
Lowe, D.G.: Distinctive Image Features from Scale-Invariant Keypoints. Internatioanl Journal of Computer Vision 60(2), 91–110 (2004)
Li, C., Yang, L., Hu, M., Liu, P.: Two Vanishing Points Error Estimation Based on Line Clustering and Condition Adjustment with Parameters. Key Engineering Materials 439-440, 1197–1202 (2010)
Ackermann, F.: Digital image correlation-performance and potential application in photogrammetry. Photogrammetric Record 11(64), 429–439 (1984)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Li, C., Liu, P., Zhou, Y. (2012). A Strategy on Corresponding Line Segment Recognition for 3D-Reconstruction. In: Deng, W. (eds) Future Control and Automation. Lecture Notes in Electrical Engineering, vol 172. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-31006-5_10
Download citation
DOI: https://doi.org/10.1007/978-3-642-31006-5_10
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-31005-8
Online ISBN: 978-3-642-31006-5
eBook Packages: EngineeringEngineering (R0)