Postprocessing for character recognition using pattern features and linguistic information

Takatoshi Yoshikawa; Masayosi Okamoto; Hiroshi Horii

doi:10.1117/12.143626

14 April 1993 Postprocessing for character recognition using pattern features and linguistic information

Takatoshi Yoshikawa, Masayosi Okamoto, Hiroshi Horii

Proceedings Volume 1906, Character Recognition Technologies; (1993) https://doi.org/10.1117/12.143626
Event: IS&T/SPIE's Symposium on Electronic Imaging: Science and Technology, 1993, San Jose, CA, United States

Abstract

We propose a new method of post-processing for character recognition using pattern features and linguistic information. This method corrects errors in the recognition of handwritten Japanese sentences containing Kanji characters. This post-process method is characterized by having two types of character recognition. Improving the accuracy of the character recognition rate of Japanese characters is made difficult by the large number of characters, and the existence of characters with similar patterns. Therefore, it is not practical for a character recognition system to recognize all characters in detail. First, this post-processing method generates a candidate character table by recognizing the simplest features of characters. Then, it selects words corresponding to the character from the candidate character table by referring to a word and grammar dictionary before selecting suitable words. If the correct character is included in the candidate character table, this process can correct an error, however, if the character is not included, it cannot correct an error. Therefore, if this method can presume a character does not exist in a candidate character table by using linguistic information (word and grammar dictionary). It then can verify a presumed character by character recognition using complex features. When this method is applied to an online character recognition system, the accuracy of character recognition improves 93.5% to 94.7%. This proved to be the case when it was used for the editorials of a Japanese newspaper (Asahi Shinbun).

Citation Download Citation

Takatoshi Yoshikawa, Masayosi Okamoto, and Hiroshi Horii "Postprocessing for character recognition using pattern features and linguistic information", Proc. SPIE 1906, Character Recognition Technologies, (14 April 1993); https://doi.org/10.1117/12.143626

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
9 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Optical character recognition

Associative arrays

Target recognition

Gold

Error analysis

Morphological analysis

Reliability

Show All Keywords

Keywords/Phrases

Search In:

Publication Years