Parameter-free geometric document layout analysis

Automatic transformation of paper documents into electronic documents requi res geometric document layout analysis at the first stage. However, variati ons in character font sizes, text line spacing, and document layout structu res have made it difficult to design a general-purpose document layout anal ysis algorithm for many years. The use of some parameters has therefore bee n unavoidable in previous methods. In this paper, we propose a parameter-fr ee method for segmenting the document images into maximal homogeneous regio ns and identifying them as texts, images, tables, and ruling lines. A pyram idal quadtree structure is constructed for multiscale analysis and a period icity measure is suggested to find a periodical attribute of text regions f or page segmentation. To obtain robust page segmentation results, a confirm ation procedure using texture analysis is applied to only ambiguous regions . Based on the proposed periodicity measure, multiscale analysis, and confi rmation procedure, we could develop a robust method for geometric document layout analysis independent of character font sizes, text line spacing, and document layout structures. The proposed method was experimented with the document database from the University of Washington and the MediaTeam Docum ent Database. The results of these tests have shown that the proposed metho d provides more accurate results than the previous ones.