Title :
Encoding of characters for OCR applications
Author :
Gaitanis, N. ; Karras, D.A. ; Perantonis, S.J.
Author_Institution :
Inst. of Inf. & Telecommun., Athens, Greece
fDate :
27 Jun- 2 Jul 1994
Abstract :
This is the first part of a systematic procedure for optical character recognition based on character features. Character feature tests result in a code assigned to the set of characters. In this paper we establish the rules for the selection of a minimal subset of features with the maximum distinguishing ability. These rules are defined by a new class of codes assigned to the set of characters so that the number of feature state differences between two characters are proportional to their degree of similarity
Keywords :
encoding; optical character recognition; OCR; character encoding; character feature tests; maximum distinguishing ability; minimal feature subset; optical character recognition; Character recognition; Classification algorithms; Encoding; Error correction; Hamming distance; Logic design; Optical character recognition software; Pattern classification; Pattern recognition; Testing;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374828
