Handwritten Bangla Numerals Recognition Based on Directional Element Feature
Yue Jing Lv · Jisuanji gongcheng · 2009
This paper presents a novel approach to automatic recognition of handwritten Bangla numerals.An effective method for Bangla numerals' feature extraction is carried out by using Directional Element Feature(DEF) combined with weighted Ending and Crossing Point(ECP) features.BP neural network is employed as a classifier.Experimental result on the numeral images obtained from real Bangladesh envelopes show that the proposed recognition method achieves a satisfied result on both recognition and reliability with the rates of 97.63% and 98.77% respectively.