Machine recognition of Hand written Characters using neural networks

Abstract

Even today in Twenty First Century Handwritten communication has its own stand and most of the times, in daily life it is globally using as means of communication and recording the information like to be shared with others. Challenges in handwritten characters recognition wholly lie in the variation and distortion of handwritten characters, since different people may use different style of handwriting, and direction to draw the same shape of the characters of their known script. This paper demonstrates the nature of handwritten characters, conversion of handwritten data into electronic data, and the neural network approach to make machine capable of recognizing hand written characters.Key Words: Machine recognition, Handwriting recognition, neural networks. INTRODUCTIONHandwritten characters are vague in nature as there may not always be sharp perfectly straight lines, and curves not necessarily be smooth, unlikely the printed characters. Furthermore, characters can be drawn in different sizes and orientation which are often supposed to be written on a baseline in an upright or downright position. Therefore, a robust handwritten recognition system can be designed by considering these limitations. However, it is quiet tedious sometimes to recognize hand written characters as it can be seen that most of the people can not even read their own written notes. Therefore, there is an obligation for a writer to write clearly. But even today in Twenty First Century Handwritten communication has its own stand and most of the times, in daily life it is globally using as means of communication and recording the information like to be shared with others.Researchers already paid many efforts in designing hand written character recognition system most of them cited as [1-5] because of its important application like bank checking process, reading postal codes and reading different forms [6]. Handwritten digit recognition is still a problem for many languages like Arabic, Farsi, Chinese, English, etc [7]. A machine can perform more tasks than a human being in the same time; this kind of application saves time and money and eliminates the requirement that a human perform such a repetitive task. For the recognition of English handwritten characters, various methods have been proposed [8][9][10][11][12]. Also a few numbers of studies have been reported for Farsi language [13][14][15]. Proposed Hand written character recognition system for machine recognition can be developed in these phases: scanning of hand written characters i.e conversion into electronic data, usually an black & white image file; some preprocessing can be applied to the image; then the feature of the character will be extracted from the image; finally, on the basis of extracted features from the image, the character can be classify to recognize using gradient descent learning method for feed forward neural network. In next sections we explore the proposed hand written character recognition system step by step. Finally, in the last section results will be...

English Scanned Document Character Recognition and Matched and Missed Matched Analysis using NN and MDA

“…LDA classical algorithm has been successfully applied and extended to various biometric signal recognition problems. The recent advancements in multilinear algebra led to a number of multilinear extensions of the LDA, Multilinear Discriminant analysis proposed for the recognition of biometric signals using their natural tonsorial representation [7]. The MDA Check Multilinear projection and maps the input data from one space to another space.…”

Handwritten Digit Recognition Using Machine Learning Algorithms

“…Experiment is carried out on NIST SD19 standard dataset. Other previous study provide the conversion of handwritten data into electronic data, nature of handwritten characters and the neural network approach to form machine that competent for recognizing hand-written characters (Perwej, 2012). The other study addresses a comprehensive criterion of handwritten digit recognition with various state of the art approaches, feature representations, and datasets (Liu, et al, 2003).…”

Reconhecimento de Caracteres Manuscritos Utilizando Sistemas Imunológicos Artificiais

“…Neste artigo os autores avaliaram o método com uma base de dados disponibilizada pelo repositório de dados UCI (University of California Irvine). No artigo [11] apresenta-se uma metodologia para o reconhecimento de caracteres manuscritos utilizando um sistema hibrido composto por duas redes neurais, sendo uma rede neural de base radial e uma rede neural multi-layer perceptron. Em [2] apresenta-se uma de neural de Kohonen com treinamento pelo algoritmo do neurônio vencedor para reconhecimento de caracteres manuscritos.…”