Cover: Real-Time Recognition of Handprinted Text

Real-Time Recognition of Handprinted Text

Published 1966

by Gabriel F. Groner


Download eBook for Free

FormatFile SizeNotes
PDF file 1.2 MB

Use Adobe Acrobat Reader version 10 or higher for the best experience.


Purchase Print Copy

 Format Price
Add to Cart Paperback59 pages $15.00

Describes a program, written in IBM 360 Assemble Language, that allows the user of an on-line computer to print data and directives on the RAND Tablet with a special pen and have them recognized and displayed immediately. The scheme recognizes 53 letters, numbers, and symbols in a wide variety of printing styles, resquiring only the usual conventions followed on coding forms. High-resolution point-by-point pen location data are gathered, displayed on the cathode ray tube screen, and analyzed while the character is being written or drawn. An average 100 data poiints per stroke (one each 4 msec) are collected, filtered, and thinned. A stroke is identified by such clues as sequence of directions, corners, and end-point location, and also by contextual clues when necessary. Multiple stroke symbols are recognized by the identification and relative location of the constituent strokes, regardless of the order in which they are written. The pen track is displayed until the character is recognized, and is then replaced by a standard hardware-generated version of the character. Previously written material remains on the display until removed. Changes, insertions, and deletions are easier than with pencil and paper. Experiments with groups of programmers, engineers, and secretaries indicate that a half-hour training period is suffcient, with 90 percent immediate recognition by the system. The scheme is in daily use in an expermental problem solving system at RAND.

This report is part of the RAND research memorandum series. The Research Memorandum was a product of RAND from 1948 to 1973 that represented working papers meant to report current results of RAND research to appropriate audiences.

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.