Character Recognition using Convolutional Neural Network
Madhuri Yadav · International Journal for Research in Applied Science and Engineering Technology · 2018
Deep learning has provided solution to many pattern recognition problems. Convolutional neural network architecture based on deep learning is proposed in this work for handwritten character recognition. The learning of Convolutional neural network (CNN) is called End-to-End learning. This paper highlights the differences between end-to-end learning and traditional learning. A CNN architecture giving an accuracy of 86.7% for Hindi handwritten characters is proposed.