Devanagri character recognition model using deep convolution neural network

In recent times, there has been a significant increase in the use of deep learning in the field of computer vision and image analysis. Deep learning is a subfield of machine learning which uses artificial neural networks that is inspired by the structure and function of the human brain. Identifying hand written text by machines has been achieved remarkable success with the use of artificial neural networks. In Optical Character Recognition for hand written text, the majority of work has been done for the popular languages such as English, Arabic or Chinese languages. There is very limited work in the literature for the handwritten character recognition for Devanagri characters. In this paper, we focus on recognition of Devanagri characters using deep convolution neural networks. Devanagri lipi is responsible for twelve languages used in India. In this paper, we optimize the network by selecting best hyperparameters for the network. Experimental results show the effectiveness of the proposed approach on the benchmark dataset.

Hali tarjima qilinmagan