A prototype document image analysis system for technical journals

Abstract

Glossary AND-OR graph (or tree). Representation of a solution strategy in which a path from the start node to the solution node requires traversing any branch at an OR node and every branch at an AND node. In a related Min-Max search used in two-person games, a path from the start node to the solution node takes the lowest cost branch at a Min node and the highest cost branch at a Max node.Bitmap. Digital representation of an image in which points are mapped to an array of binary pixels. Branch-and-bound.A search technique that avoids paths certain to lead to higher cost solutions than the best solution obtained so far.

Document Image Processing for Hospital Information Systems

“…Studies and research for document image analysis systems have been reported [23]- [31]. As related works to ruled line extraction, the detection methods using the Hough transform technique are reported by literature [23]- [27].…”

Enriching Historical Manuscripts: The Bovary Project

“…Numerous methods using one of these strategies have been proposed for the analysis of machine printed documents. Among the most popular we can cite Kise's method [13] based on area Voronoi diagram, O'Gorman's Docstrum method [14] based on neighbor clustering and Nagy's X-Y cut [15] based on the analysis of projection profiles. These methods provide good results on printed documents, but are not directly adapted to handwritten documents, because they generally take only into account global features of the page, and are thus dedicated to well structured documents.…”

Transforming paper documents into XML format with WISDOM++

“…In addition, we note that knowledge used in top-down approaches is typically derived from the relations between the geometric and the logical structures of specific classes of documents. This is the case of page grammars (Nagy et al 1992) and geometric trees (Dengel and Barth 1988), which are used to segment document images and simultaneously associate some layout components with the logical structure. In WISDOM++ this class-specific knowledge is solely required in the document classification and understanding steps and it is automatically learned from examples of documents, as explained in the next section.…”

“…Typically such rules are handcoded for particular classes of documents (Nagy et al 1992), requiring fine-tuning and great human effort. In WISDOM++ rules are automatically generated by means of machine learning algorithms that induce them from a set of training examples, for which the final user has already defined the correct class and has specified the layout components with a logical meaning (logical components) (Esposito et al 1999).…”