Abstract
The structure recognition of bar-graphs is one of interesting subjects in the research fields of document understanding and drawing interpretation. In this paper, we address the recognition method of bar-graphs under the assumption that bar-graph has its own layout structure. The bar-graphs are very simple in comparison with other graphs, but the structures are too complicated: we observe various structures or many different representations in our environment. Our method is constructed flexibly so as to analyze such complicated structures successfully: our layout network, which represents the knowledge of layout structure of bar-graphs, is composed of three subnets with a view to establishing the concise specification of variant and invariant structures.
Preview
Unable to display preview. Download preview PDF.
References
R. Kasturi and et al.: “A System for Recognition and Description of Graphics”, Proc. of ICPR'88, Vol. 1, pp. 255–259 (1988).
R. P. Futrelle and et al.: “Understanding Diagrams in Technical Documents”, IEEE Computer, Vol. 25, No. 7, pp. 75–78 (1992).
Y. Nakamura and R. Furukawa: “Integration of Pattern Information and Natural Language Information toward Diagram Understanding”, Trans. on IPSJ, Vol. 36, No. 1, pp. 196–205 (1995) [in Japanese].
M. H. Lee, N. Babaguchi and T. Kitahashi: “Symbolization and Presentation of Graph Images for Intelligent Communication of Document Images”, Proc. of ACCV'95, Vol. 3, pp. 680–684 (1995).
N. Yokokura and T. Watanabe: “Recognition of Composite Elements in Bar Graphs”, Proc. of MVA'96, pp. 348–351 (1996).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yokokura, N., Watanabe, T. (1997). Recognition of various bar-graph structures based on layout model. In: Chin, R., Pong, TC. (eds) Computer Vision — ACCV'98. ACCV 1998. Lecture Notes in Computer Science, vol 1352. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63931-4_205
Download citation
DOI: https://doi.org/10.1007/3-540-63931-4_205
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63931-2
Online ISBN: 978-3-540-69670-4
eBook Packages: Springer Book Archive