Title :
A new algorithm for extracting character feature points
Author :
Cao, Zhonghua ; Dai, Liping ; Wang, Yan
Author_Institution :
Sch. of Software, Jiangxi Normal Univ., Nanchang, China
Abstract :
A new algorithm for extracting character feature points is put forward in this paper. It first get the number of background regions and the corresponding features through analyzing the discrete circle template segmented by character skeleton, and then obtain the character feature points, finally determine the location of fork point accurately by using the maximum circle criterion. Experimental results show that this new algorithm could extract the feature points in a fast, accurate and robust way.
Keywords :
character recognition; feature extraction; character feature point extraction; character skeleton; discrete circle template; fork point; maximum circle criterion; Character recognition; Data mining; Feature extraction; Handwriting recognition; Robustness; Text recognition; Rutoviz crossing number; circle template; fork point;
Conference_Titel :
Multimedia Technology (ICMT), 2011 International Conference on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-61284-771-9
DOI :
10.1109/ICMT.2011.6001969