HANDWRITTEN DEVANAGARI CHARACTER RECOGNITION USING DEEP LEARNING - CONVOLUTIONAL NEURAL NETWORK (CNN) MODEL

Abstract

Handwritten character recognition is increasingly important in a variety of automation fields, for
example, authentication of bank signatures, identification of ZIP codes on letter addresses, and
forensic evidence, etc. Handwritten character recognition is the process where the machine detects
and recognizes the characters from a text image and converts that processed data into a code which is
understood by the machine. It is a fundamental yet challenging task in the field of pattern recognition.
In this paper, we used a new public image dataset for Devnagari script character: Devnagari Character
Dataset (DCD). This considered dataset consists of 92 thousand images of 46 different classes of
characters of Devnagari script segmented from handwritten documents. This paper also explore the
challenges in recognition of Devnagari characters. Along with the dataset, a deep learning based
convolutional neural network (CNN) architecture is proposed in this paper for recognition of those
handwritten characters in an unrestricted environment. Deep Convolutional Neural Network have
shown superior results to traditional shallow networks in many recognition tasks. Keeping distance
with the regular approach of character recognition by Deep CNN, we focus the use of Dropout and
dataset increment approach to improve test accuracy. By implementing these techniques in Deep
CNN, we were able to increase test accuracy by nearly 0.98 percent. The proposed architecture scored
highest test accuracy of 98.13% on theconsidered dataset. The results indicate that the proposed
model may be a strong candidate for handwritten character recognition and automated handwritten
Devnagari script character recognition applications.