Document recognition for facsimile transmission

A file containing a text document is much more useful than a raw fax. The information it contains is at a much higher level of representation than is the case for the fax image. It requires much less work in the sense of the application of computing power to make use of the information. Taking a document produced using a word processing program and sending it as a fax is therefore an act of supreme folly in which much information is lost, the high level representation is wantonly discarded, and the file size is multiplied by two orders of magnitude. The objective of this research into document recognition is to therefore to reclaim the high level information to allow the document to be routed, indexed, integrated into a digital document workflow system, or best of all to allow it to be simply discarded. This is a transformation from a large volume of low quality information to a smaller volume of high quality information