Optimal Line and Arc Detection on Run-Length Representations

The robust detection of lines and arcs in scanned documents or technical drawings is an important problem in document image understanding. We present a new solution to this problem that works directly on run-length encoded data. The method finds globally optimal solutions to parameterized thick line and arc models. Line thickness is part of the model and directly used during the matching process. Unlike previous approaches, it does not require any thinning or other preprocessing steps, no computation of the line adjacency graphs, and no heuristics. Furthermore, the only search-related parameter that needs to be specified is the desired numerical accuracy of the solution. The method is based on a branch-and-bound approach for the globally optimal detection of these geometric primitives using runs of black pixels in a bi-level image. We present qualitative and quantitative results of the algorithm on images used in the 2003 and 2005 GREC arc segmentation contests.

[1]  Frédéric Jurie,et al.  Solution of the Simultaneous Pose and Correspondence Problem Using Gaussian Error Model , 1999, Comput. Vis. Image Underst..

[2]  Thomas M. Breuel,et al.  Fast recognition using adaptive subdivisions of transformation space , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[3]  Thomas M. Breuel.  On the use of interval arithmetic in geometric branch and bound algorithms , 2003, Pattern Recognit. Lett..

[4]  David M. Mount,et al.  Ecient Algorithms for Robust Feature Matching , 1998 .

[5]  Wenyin Liu.  Report of the Arc Segmentation Contest , 2003, GREC.

[6]  Dov Dori,et al.  Sparse Pixel Vectorization: An Algorithm and Its Performance Evaluation , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Jean Ponce,et al.  Computer Vision: A Modern Approach , 2002 .

[8]  M. H. van Emden,et al.  Interval arithmetic: From principles to implementation , 2001, JACM.

[9]  Remco C. Veltkamp,et al.  Reliable and Efficient Pattern Matching Using an Affine Invariant Metric , 1999, International Journal of Computer Vision.

[10]  Theo Pavlidis,et al.  Algorithms for Graphics and Imag , 1983 .

[11]  Luc Jaulin,et al.  Applied Interval Analysis , 2001, Springer London.

[12]  Karl Tombre,et al.  Improving arc detection in graphics recognition , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[13]  Dov Dori,et al.  Extended Summary of the Arc Segmentation Contest , 2001, GREC.

[14]  Richard O. Duda,et al.  Use of the Hough transformation to detect lines and curves in pictures , 1972, CACM.

[15]  Dave Elliman,et al.  TIF2VEC, An Algorithm for Arc Segmentation in Engineering Drawings , 2001, GREC.

[16]  Thomas M. Breuel.  Robust least-square-baseline finding using a branch and bound algorithm , 2001, IS&T/SPIE Electronic Imaging.

[17]  Clark F. Olson,et al.  Locating geometric primitives by pruning the parameter space , 2001, Pattern Recognit..

[18]  Thomas M. Breuel.  Representations and metrics for off-line handwriting segmentation , 2002, Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition.

[19]  Daniel P. Huttenlocher,et al.  Comparing Images Using the Hausdorff Distance , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  David M. Mount,et al.  Efficient algorithms for robust feature matching , 1999, Pattern Recognit..

[21]  Thomas M. Breuel,et al.  Finding lines under bounded error , 1996, Pattern Recognit..

[22]  T. Pavlidis.  Algorithms for Graphics and Image Processing , 1981, Springer Berlin Heidelberg.

[23]  Xavier Hilaire.  RANVEC and the Arc Segmentation Contest , 2001, GREC.