DocumentCode :
153369
Title :
Performance Improvement in Local Feature Based Camera-Captured Character Recognition
Author :
Matsuda, Tadamitsu ; Iwamura, Mikio ; Kise, Kenji
Author_Institution :
Dept. of CSIS, Osaka Prefecture Univ., Sakai, Japan
fYear :
2014
fDate :
7-10 April 2014
Firstpage :
196
Lastpage :
201
Abstract :
Concerning camera-captured Japanese character recognition, we have proposed a method to recognize characters, both simple and complex, that may not be linearly aligned and may be printed with a complex background. Recognition is performed based on local features and their arrangement. The arrangement is validated with an algorithm called local RANSAC. However, at least four corresponding local features are required. To relax that condition, we propose a new recognition method making it possible to recognize a character region with at least three corresponding local features. This method enables recall and precision to be improved with the simpler characters using more corresponding local features and computation times to be reduced by 7%.
Keywords :
character recognition; object recognition; random processes; camera-captured Japanese character recognition; local RANSAC; local feature; Character recognition; Databases; Feature extraction; Image recognition; Robustness; Text recognition; Vectors; RANSAC; affine transformation matrix; local feature; reference point; scene character recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Document Analysis Systems (DAS), 2014 11th IAPR International Workshop on
Conference_Location :
Tours
Print_ISBN :
978-1-4799-3243-6
Type :
conf
DOI :
10.1109/DAS.2014.78
Filename :
6830997
Link To Document :
بازگشت