ADAPT: Automated Document Architecture Processing and Tagging

Abstract

For three years, Project ADAPT has investigated methods that must be developed to support economical conversion of information into electronic information systems-not just symbol recognition but document recognition. ADAPT system software provides a prototype methodology for hands-off conversion of document images to a tagged, structured document database. ADAPT also includes pilot efforts to increase optical character recognition (OCR) accuracy and to increase automated structure analysis capabilities of systems-document format recognition. Digitization of high-quality microfilm material may be the most practical way to load many documents into a conversion system, considering materials han..