Abstract
This paper presents an algorithm for extracting lines from hand drawings. It starts from contour pixel tracing, fits them into contour segments, and then extracts skeleton lines from the contour segments. The algorithm finds all contours in one scan of the input matrix without detecting and marking multiple pixels. In line extraction, the method Elastic Contour Segment Tracing is proposed which extracts lines by referring to the contour segments at both sides, overcoming noise and passing through blotted areas by fitting and extrapolation.
Experiments on free hand mechanical drawings, sketches, letter/numerals, as well as Chinese characters are carried out and satisfactory results are achieved.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.References
Naccache N J, Shinghal R. SPTA: a proposed algorithm for thinning binary patterns.IEEE Trans. on System, Man and Cybernetics, 1984, 14(3): 409–418.
Chuei Net al. New algorithm for thinning binary images and Chinese characters.Computer Processing of Chinese & Oriental Languages, 1986, 2(3): 169–179.
Okazaki Aet al. An automatic circuit diagram reader with loop-structure based symbol recognition.IEEE Trans. on Pattern Analysis and Machine Intelligence, 1988, 10(3): 331–341.
Sakakura Tet al. A hierarchical analysis of Chinese characters and their graph representation. In:Proc. 1982 Int’l Conf. of Chinese Language Computing Society, 1982, pp. 310–321.
Legauti R, Suen C Y. Contour tracing and pattern approximations for digital patterns. In:Computer Vision and Shape Recognition, Krzyzak Aet al. eds., World Scientific, 1989, pp. 225–259.
Mori Set al. Line fitting and its application to stroke segmentation of hand printed Chinese characters. In:Proc. 7th Int’l Conf. on Pattern Recognition, 1984, pp. 366–369.
Li B, Suen C Y. A knowledge-based thinning algorithm.Pattern Recognition, 1991, 24(12): 1211–1221.
Xia Y. Minimizing the computing complexity of iterative sequential thinning algorithm. In:Proc. of 9th Int’l Conf. on Pattern Recognition, 1988, pp. 721–723.
Chen Y, Hsu W. 1-subcycle parallel thinning algorithm for producing perfect 8-curves and obtaining isotropic skeleton of an L-shape pattern. In:Proc. of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1989, pp. 208–215.
Pavlidis T. Algorithm for Graphics and Image Processing. Springer-Verlag, 1982.
Weiss Isaac. Line fitting in a noisy image.IEEE Trans. on Pattern Analysis and Machine Intelligence, 1989, 11(3): 325–328.
Califano Aet al. Generalized neighborhoods: a new approach to complex parameter feature extraction. In:Proc. of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 1989, pp. 192–199.
Rosin P L, West G. Segmentation of edges into lines and arcs.Image and Vision Computing, 1989, 7(2): 109–114.
Suzuki S. Graph-based vectorization method for line patterns. In:Proc. of Computer Society Conference on Computer Vision and Pattern Recognition, 1988, pp. 616–621.
Zhao M. Two-dimensional extended attribute grammar method for the recognition of hand printed Chinese characters.Pattern Recognition, 23(7):685–95.
Zhao M. Automatic three-dimensional object recovery from three-view drawings. In:Proc. of Int’l Conf. on Information Engineering, Singapore, 1991.
Author information
Authors and Affiliations
Additional information
This paper is supported by National Natural Science Foundation of China and Research Grant of Asian Institute of Technology, Bangkok, Thailand.
Zhao Ming received his Ph.D. degree from The Institute of Software, The Chinese Academy of Sciences in 1989. He received his M. S. degree from the Institute of Computing Technology, The Chinese Academy of Sciences in 1984 and his B.S. degree from the Mathematics Department, Anhui University in 1982, both in computer science.
At present he is a senior scientist with the Research Laboratories, Telecom Australia. His present research interests include the application of expert systems in telecommunications, distributed artificial intelligence, knowledge acquisition, machine learning, software engineering methodologies and human computer interactions.
Rights and permissions
About this article
Cite this article
Zhao, M. A line extraction algorithm for hand drawings. J. of Comput. Sci. & Technol. 10, 2–14 (1995). https://doi.org/10.1007/BF02939517
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF02939517