A DOCUMENT RECOGNITION SYSTEM AND ITS APPLICATIONS

This paper describes a document entry system called the Document Recog nition System (DRS), which facilitates the conversion of printed docum ents into electronic form. DRS was developed on a personal computer (P C) with an adapter card for recognizing more than 3000 Kanji character s. It provides a flexible framework for object-oriented management of data and processing modules. The framework allows the user to change t he combination of processing modules and to select pipelining (paralle l processing) or sequential processing. DRS includes processing module s for layout analysis functions such as blob detection, block segmenta tion, and model matching, and for character recognition functions such as Kanji character recognition, Japanese postprocessing, postprocessi ng by a user, and error correction through a user interface. The chara cter recognition functions on the card and the other processing-relate d recognition functions on the PC work cooperatively in the proposed f ramework. Within the basic framework, we have customized DRS for pract ical applications. Examples of successful applications-entry into a te xt database, creation of an electronic catalog, entry of family regist ration data, and entry of tag data in a manufacturing process-provide evidence of the processing accuracy and robustness of the framework.