Image analysis – Learning systems – Trainable classifiers or pattern recognizers
Patent
1994-04-05
1995-12-26
Boudreau, Leo H.
Image analysis
Learning systems
Trainable classifiers or pattern recognizers
382203, G06K 962
Patent
active
054795338
ABSTRACT:
Pattern recognition method and apparatus are provided which can recognize not only standard typewritten characters but also handwritten characters and various styles of fonts without the necessity of learning, formation of rules, and complicated algorithm for extracting characteristics of patterns which have been required to conventional pattern recognition apparatus. A signal representing a pattern to be recognized is inputted and converted into binary signals corresponding to the pattern. The binary signals are compared with predetermined reference patterns having indefinite boundaries. The reference patterns each are formed based on one or more multi-dimensional membership functions. A certainty factor of the pattern recognition is calculated from the comparison as the pattern recognition result.
REFERENCES:
patent: 4177448 (1979-12-01), Brayton
patent: 4805224 (1989-02-01), Koezuka et al.
patent: 5121441 (1992-06-01), Chefalas et al.
patent: 5150425 (1992-09-01), Martin et al.
patent: 5222155 (1993-06-01), Delanoy et al.
Takata, K. "Signature Identification System Using a Fuzzy Template" Nov. 1991 abstract.
Shichang, Wu "A fuzzy pattern recognition method of hand printed characters" 1986 abstract.
Boudreau Leo H.
Kelley Christopher
Yamatake-Honeywell Co. Ltd.
LandOfFree
Pattern recognition apparatus and method using fuzzy logic does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Pattern recognition apparatus and method using fuzzy logic, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Pattern recognition apparatus and method using fuzzy logic will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1375260