Image analysis – Pattern recognition – Feature extraction
Reexamination Certificate
1999-06-24
2003-11-04
Mariam, Daniel G. (Department: 2621)
Image analysis
Pattern recognition
Feature extraction
C382S179000, C382S181000, C382S185000, C382S190000, C382S229000
Reexamination Certificate
active
06643401
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
Recently, the demand for advanced character recognition technology for peripheral input devices has become very strong. The present invention relates to an apparatus and a method of recognizing a character, and a storage medium for storing a character recognition program. Specifically, the present invention relates to an apparatus and a method of recognizing characters by distinguishing hand-written characters from typed characters in a document to more precisely recognize the characters, and a storage medium for storing the character recognition program.
2. Description of the Related Art
To recognize a document containing both hand-written characters and typed characters in the conventional technology, two processes, that is, a process of recognizing hand-written characters and a process of recognizing typed characters, are first performed. Then, the two process results are compared with each other to obtain a recognition result of higher reliability, thereby recognizing the document.
The above described conventional technology obtains a recognition result of high reliability by performing two processes (separately recognizing hand-written characters and typed characters), and then comparing the recognition results of the two processes. Therefore, there has been the problem with the conventional technology that it takes a long time to perform the recognizing processes.
SUMMARY OF THE INVENTION
The present invention has been developed to solve the above described problem with the conventional technology, and aims at recognizing both hand-written characters and typed characters at high precision, and shortening the time required to perform the character recognizing process.
To solve the above described problem, the present invention recognizes the features of a pattern, performs a process of recognizing hand-written characters when the features of the pattern imply hand-written characters, and performs a process of recognizing typed characters when the feature of the pattern imply typed characters.
Thus, even when a document contains both hand-written characters and typed characters, a precise recognition result can be obtained only by performing only one of the hand-written characters recognizing process and the typed characters recognizing process, thereby successfully shortening the time required to perform the character recognizing process without lowering the precision in recognizing characters.
According to an aspect of the present invention, a hand-written character or a typed character can be discriminated based on the form or the regularity of a pattern.
Since typed characters are more regular in pattern shape and arrangement than hand-written characters, hand-written characters can be distinguished from typed characters by checking the regularity in pattern shape and arrangement, thereby shortening the time taken for performing the subsequent processes.
REFERENCES:
patent: 4259661 (1981-03-01), Todd
patent: 4516262 (1985-05-01), Sakurai
patent: 4700400 (1987-10-01), Ross
patent: 4910787 (1990-03-01), Umeda et al.
patent: 4998626 (1991-03-01), Ota
patent: 5025475 (1991-06-01), Okabe
patent: 5521985 (1996-05-01), Camp, Jr. et al.
patent: 5535287 (1996-07-01), Niki
patent: 5561720 (1996-10-01), Lellmann et al.
patent: 5581633 (1996-12-01), Hotta et al.
patent: 5872864 (1999-02-01), Imade et al.
patent: 60-118985 (1985-06-01), None
patent: 63-79193 (1988-04-01), None
patent: 63-298494 (1988-06-01), None
patent: 10-154204 (1998-06-01), None
patent: 10-162102 (1998-06-01), None
patent: 10-162103 (1998-06-01), None
Kuhnke, et al “A system for machine-written and hand-written character distinction”, IEEE, pp. 811-814, 1995.*
Violante, et al discloses “A Computationally Efficient Technique for Discriminating between Hand-written and Printed Text”, IEEE, 1995.
Kashioka Junji
Naoi Satoshi
Fujitsu Limited
Mariam Daniel G.
Staas & Halsey , LLP
LandOfFree
Apparatus and method for recognizing character does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for recognizing character, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for recognizing character will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3140674