Development of Handwritten Numeral String Recognition Algorithm

Abstract

In this today era, every one busy with the fullest and with the increasing digitization everyone want to complete their work easily and as fast as possible. If it was possible to give input to a digital machine in our own handwritten and an in our own language, it would be very easy and probably less time consuming. Researcher are continuously working on handwritten recognizer. One of the basic step of handwritten recognizer is the recognition of handwritten numerals. Handwritten recognition can done in two way, namely offline and online. The step involve in this process is collect the data base, pre-processing of image, segmentation, features extraction and finally classification or recognition the numerals.

This report going to talk about the handwritten Hindi and Gujarati Numerals. In the first part, am going to talk about the optical character recognition (OCR) system and decide which features we are going to use for the Hindi and Guajarati Numerals recognition .In the second part, am going to talk about the features extraction method base on the Zoning and projection profile method. In the next part am highlight the which classifier am going to use for the features extraction method.