A description of an elementary procedure for the synthesis of a character-recognition device based on a learning experiment. Using information derived from a significant sample of the set of characters to be read and given identification of the samples by a human operator, a computer defines a set of "filters." These filters may then be used to transform unknown characters having similar type characteristics. During the recognition process a probability matrix for each character in the alphabet is used to compute a figure of merit for the hypothesis that an unknown character is the same as a known character. It is shown that this elementary model may aid in constructing a fast input device for a language translation machine if it was able to make use of frequency distribution characteristics of the dictionary. A possible implementation with a raw character reading rate up to 500 characters a second appears feasible.
This report is part of the RAND Corporation paper series. The paper was a product of the RAND Corporation from 1948 to 2003 that captured speeches, memorials, and derivative research, usually prepared on authors' own time and meant to be the scholarly or scientific contribution of individual authors to their professional fields. Papers were less formal than reports and did not require rigorous peer review.
Permission is given to duplicate this electronic document for personal use only, as long as it is unaltered and complete. Copies may not be duplicated for commercial purposes. Unauthorized posting of RAND PDFs to a non-RAND Web site is prohibited. RAND PDFs are protected under copyright law. For information on reprint and linking permissions, please visit the RAND Permissions page.
The RAND Corporation 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.