Automated detection and segmentation of table of contents page and index pages from document images

A Complete System for Detection and Identification of Tabular Structures from Document Images

“…As a result the identification/segmentation of the TOC and Index page from the scanned document has attracted researchers [14,15] to put forward a couple of schemes to do the same.…”

OCR-Free Table of Contents Detection in Urdu Books

“…Despite the presence of some recent developments in layout analysis systems for Arabic and Urdu documents [9], the non-existence of commercial or opensource OCR techniques for these scripts make it difficult to navigate efficiently through scanned documents. Even the OCR-free technique presented by [7] can not be applied to these scripts due to highly non-uniform distribution of intra and inter word distances [10]. Moreover, lack of knowledge about location of the digits would make it impossible to differentiate between a ToC page and a page whose structure is similar to a ToC page.…”

“…The distribution of digits are then investigated in the second stage for presence of a column structure by combining vertical projection profile analysis with runlength encoding. So, the present work may be seen as an improvement of [7] as it uses the knowledge of digits' location in a document as the criterion for ToC detection.…”

“…Gao et al [2] proposed a ToC parsing and decorative elements detection technique based on clustering. Mandal et al [7] used the spatial distribution properties of contents for detection of ToC pages. They identified ToC pages based on analysis of rightmost words of each line, where words were isolated in a text line based on intraword spaces.…”

“…All techniques mentioned above for detection of ToC pages (except [7]) share the use of textual similarity (OCR) to detect a ToC page. The use of OCR definitely increases the detection probability of a ToC page.…”