METHOD AND APPARATUS FOR DETERMINING THE FREQUENCY OF WORDS IN A DOCUMENT WITHOUT DOCUMENT IMAGE DECODING

摘要

A method and apparatus for determining wordfrequency from a document without first converting thedocument to character codes. The method includesmorphological image processing to determine word unitcharacteristics for placement into equivalence classes utilizingnon-content based information. Word shape representationsare preferably determined and compared to defineequivalent word units.