Example-Based Logical Labeling of Document Title Page Images

This paper presents a flexible and effective example- based approach for labeling title pages which can be used for automated extraction of bibliographic data. The labels of interest are "title", "author", "abstract" and "affiliation". The method takes a set of labeled document layouts and a single unlabeled document layout as input and finds the best matching layout in the set. The labels of this layout are used to label the new layout. The similarity measure for layouts combines structural layout similarity and textural similarity on the block-level. Experimental results yield accuracy rates from 94.8% to 99.6% obtained on the publicly available MARG dataset. This shows that our lightweight method has equivalent and partially better performance when compared to other more complex labeling methods known from the literature.

[1]  Azriel Rosenfeld,et al.  Document structure analysis algorithms: a literature survey , 2003, IS&T/SPIE Electronic Imaging.

[2]  Motoi Iwata,et al.  Segmentation of Page Images Using the Area Voronoi Diagram , 1998, Comput. Vis. Image Underst..

[3]  Daniel X. Le,et al.  Automated labeling in document images , 2000, IS&T/SPIE Electronic Imaging.

[4]  Marco Aiello,et al.  Document understanding for a broad class of documents , 2002, Int. J. Document Anal. Recognit..

[5]  Thomas M. Breuel,et al.  Distance measures for layout-based document image retrieval , 2006, Second International Conference on Document Image Analysis for Libraries (DIAL'06).

[6]  Thomas M. Breuel,et al.  Document image zone classification - a simple high-performance approach , 2007, VISAPP.

[7]  R. A. Leibler,et al.  On Information and Sufficiency , 1951 .

[8]  Song Mao.  BAYESIAN LEARNING OF 2D DOCUMENT LAYOUT MODELS FOR PRESERVATION METADATA EXTRACTION , 2004 .

[9]  Song Mao,et al.  Style-independent document labeling: design and performance evaluation , 2003, IS&T/SPIE Electronic Imaging.

[10]  Joachim M. Buhmann,et al.  Empirical evaluation of dissimilarity measures for color and texture , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[11]  David S. Doermann,et al.  Logical Labeling of Document Images Using Layout Graph Matching with Adaptive Learning , 2002, Document Analysis Systems.