A NEURAL NETWORK METHOD TO PERFORM OPTICAL CHARACTER RECOGNITION OF PRINTED CHARACTERS

Abstract

Text recognition has gained a lot of attention in the last few years. The Optical character recognition is one of the method used for text recognition .Also printed document with different font and also some low resolution images . This paper describes all the method used and also explanation for non-horizontal text which are used and explanation of historical of the text from old documents which have been digitalized. The Numerals and letters are recognized by Convolutional Neural Networks The results obtained by Convolutional Neural Networks seems to be promising in recognizing the characters in comparison to results obtained through other algorithms such as Support Vector Machines and Artificial Neural Networks. Characters are recognized by using Convolutional Neural Networks. The results obtained by Convolutional Neural Networks seem promising when it comes to detecting signs compared to the results obtained by other machine learning algorithms such as support vector machines and artificial neural networks.