Abstract.
When comparing document images based on visual similarity it is difficult to determine the correct scale and features for document representation. We report on a new form of multivariate granulometries based on rectangles of varying size and aspect ratio. These rectangular granulometries are used to probe the layout structure of document images, and the rectangular size distributions derived from them are used as descriptors for document images. Feature selection is used to reduce the dimensionality and redundancy of the size distributions while preserving the essence of the visual appearance of a document. Experimental results indicate that rectangular size distributions are an effective way to characterize visual similarity of document images and provide insightful interpretation of classification and retrieval results in the original image space rather than the abstract feature space.
Similar content being viewed by others
References
Antonacopoulos A (1998) Page segmentation using the description of the background. Comput Vision Image Understand 70(3):350-369
Batman S, Dougherty ER, Sand F (2000) Heterogeneous morphological granulometries. Patt Recog 33:1047-1057
Breuel T (2000) Layout analysis by exploring the space of segmentation parameters. In: Proceedings of the 4th international workshop on document analysis systems (DAS’2000), Rio de Janeiro, 10-13 December 2000
Chandler D (2001) Semiotics: the basics. Routledge, London
Doermann DS (1998) The indexing and retrieval of document images: a survey. Comput Vision Image Understand 70(3):287-298
Dougherty ER, Pelz J, Sand F, Lent A (1992) Morphological image segmentation by local granulometric size distributions. J Electron Imag 1:46-60
Haralick RM, Katz PL, Dougherty ER (1995) Model-based morphology: the opening spectrum. Graph Models Image Process 57(1):1-12
Maragos P (1989) Pattern spectrum and multiscale shape representation. IEEE Trans Patt Analysis Mach Intell 11:701-716
Matheron G (1975) Random sets and integral geometry. Wiley, New York
Serra J (1982) Image analysis and mathematical morphology. Academic, New York
Shin CK, Doermann DS (2000) Classification of document page images based on visual similarity of layout structures. In: Proceedings of SPIE Document Recognition and Retrieval VII, San Jose, 26-27 January 2000, pp 182-190
Vincent L (2000) Granulometries and opening trees. Fundamenta Informatica 41(1-2):57-90
Author information
Authors and Affiliations
Corresponding author
Additional information
Received: 16 November 2002, Accepted: 20 June 2003, Published online: 12 September 2003
Correpondence to: Andrew D. Bagdanov
Rights and permissions
About this article
Cite this article
Bagdanov, A.D., Worring, M. Multiscale document description using rectangular granulometries. IJDAR 6, 181–191 (2003). https://doi.org/10.1007/s10032-003-0112-1
Issue Date:
DOI: https://doi.org/10.1007/s10032-003-0112-1