Abstract A character pattern to be recognized is divided into a predetermined number of parallel segments of equal width extending in one direction and character information on respective ones of the segments is read out via a line of photoelectric cells. Two pieces of the character information on two adjacent segments are evaluated together in developing symbolic information indicative of the correlation between two adjacent segment patterns. By the action of simplified feature extraction systems, a plurality of two-dimensional features are established in accordance with all combinations of the symbolic information which have been derived throughout the scanning procedure. The symbolic information is developed particularly with regard to the number of loops within the character pattern, the shape of the left edge portion, the shape of the right edge portion, the shape of the upper edge portion and the shape of the lower edge portion of the character pattern. These combinations of the symbolic information are compared with reference information prestored in a memory of an OCR (optical character reader).
An optical character reader comprising: memory means for storing a character pattern in addressable parallel segments, all of said segments extending in a common linear direction; means for retrieving character pattern information from said addressable parallel segments; classification means responsive to said means for retrieving for comparing character pattern information in retrieved adjacent segments two at a time and classifying the combined character pattern information of each two retrieved adjacent segments into a predetermined number of groups of predefined shapes without regard to specific address information within each of said segments; and identification means for comparing the combinations of groups of predefined shapes of the respective adjacent segments with reference information to identify the character patterns, said reference information including predetermined combinations of said groups of predefined shapes for each character to be identified.
Claim 2
Dependent Claim
134 words
An optical character reader comprising: memory means for storing a character pattern in addressable parallel segments, all of said segments extending in a common linear direction; means for retrieving character pattern information from said addressable parallel segments; classification means responsive to said means for retrieving for comparing character pattern information in retrieved adjacent segments two at a time and classifying the combined character pattern information of each two retrieved adjacent segments into a predetermined number of groups of predefined shapes without regard to specific address information within each of said segments; and identification means for comparing the combinations of groups of predefined shapes of the respective adjacent segments with reference information to identify the character patterns, said reference information including predetermined combinations of said groups of predefined shapes for each character to be identified.
Claim 3
Dependent Claim
145 words
The optical character reader of claim 1, wherein said classification means comprises: means for correlating said predefined shapes of the character pattern information in the respective adjacent segments to at least the following features of the overall character pattern; (a) the number of loops in said character pattern, (b) the shape of the lower edge portion of said character pattern, (c) the shape of the upper edge portion of said character pattern, (d) the number of connections of lateral lines within said character pattern, (e) the number of hollows on the left edge portion of said character pattern, (f) the number of hollows on the right edge portion of said character pattern, and (g) the number of swells on the right edge portion of said character pattern, whereby said reference information is correlated to the same said features to facilitate identification of said character patterns.