Browse Prior Art Database

Weighted Template Match OCR Classification

IP.com Disclosure Number: IPCOM000068220D
Original Publication Date: 1979-Dec-01
Included in the Prior Art Database: 2005-Feb-20

Publishing Venue

IBM

Related People

Authors:
Goddard, RD Cook, WP [+details]

Abstract

The weighted template match algorithm described hereinafter is an effective classifier for identifying optically scanned characters. In conventional array scanners, a light source illuminates an original document positioned on the document platen of a scanner. The reflected light from the document is focused onto a diode array which detects the light levels across the scanned page. The photodiode array output is usually quantized into 16 levels and stored as binary information (4 bits per diode). By resetting the diodes and moving the paper with respect to the array, all of the reflective information is stored digitally.