Handwritten Bangla Character Recognition Using Normalized Cross Correlation
Md. Azher Uddin · IOSR Journal of Computer Engineering · 2014
This paper presents a system for recognizing Bangla handwritten characters using template matching algorithm based on normalized cross-correlation.The Bangla alphabets Sorborno and Banjonborno have been chosen to test the recognition System.Character is recognized by analyzing its shape and comparing its features that differentiates each character.The required images are acquired using a digital camera.The color images are then cropped, resized, and converted to binary images.Finally, Template Matching algorithm based on normalized cross-correlation technique is used for recognition of candidate regions.Experimental results show the relatively high accuracy of the developed method.The work has valuable application in the development of an OCR system for handwritten Bangla text.