Title :
Recognition of bangla characters using regional features and principal component analysis
Author :
Naser, M.A. ; Hossain, Md Murad ; Tito, Shafiqur Rahman ; Hoque, Md Ashraful
Author_Institution :
Dept. of Electr. & Electron. Eng., Islamic Univ. of Technol., Gazipur, Bangladesh
Abstract :
Feature extraction is the most important step of character recognition. Accuracy of the recognition mostly depends on it. Ten regional features are proposed in this paper. Principal component analysis is used to reduce the data set as well as to remove redundancy. For recognition, K-nearest neighbor classifier is used and that yields a very high rate of 100%.
Keywords :
character recognition; feature extraction; image classification; principal component analysis; Bangla character recognition; k-nearest neighbor classifier; principal component analysis; regional feature extraction; K-nearest neighbor classifier; Principal component analysis; Regional features;
Conference_Titel :
Electrical and Computer Engineering (ICECE), 2010 International Conference on
Conference_Location :
Dhaka
Print_ISBN :
978-1-4244-6277-3
DOI :
10.1109/ICELCE.2010.5700740
