An Approach to Automated Contextual Character Recognition

摘要

The purpose of this paper is to combine current knowledge and technology to set forth an approach for designing a machine capable of recognizing various styles of typed characters. This approach differs considerably from existing approaches in that it has the capability to make use of context to aid in recognition. This is accomplished by the tabulation of the probability that one particular letter follows another letter. Hence, when attempting to recognize a typed character, the machine has knowledge of the previous letters or letter that it recognized, and can make use of this information in identifying the typed character presently being processed.