Image-based Structure analysis for a Table of Contents and Conversion to XML Documents

In this paper, we describe a method of image-based structure analysis and conversion into XML documents for a table of contents in a book. In the table of contents, structural elements such as chapter, section are featured by the extent of indention and the size of character. We focus on them, and classify the text lines into some groups using the two features of text line. We relate the groups with the structure tree (DTD in XML) prepared in advance. Our system can convert a table of contents on a paper into the XML documents automatically using the structure tree.