Mongolian Handwriting Character Recognition Based On Convolutional Neural Network(CNN)

DOI

10.2991/icimm-16.2016.105How to use a DOI?

Keywords

Convolutional neural network, LeNet-5, Mongolian Cyrillic alphabet

Abstract

The ability of multilayer networks with increasing complexity, multidimensional and nonlinear mappings from large collections of examples can be good for image recognition tasks. In this case, Conventional neural networks(CNN) is compatible for image recognition task as it contains multiple layers of small neuron collection. In this paper, we accomplish Mongolian character recognition by using algorithm of Convolutional neural network. The input of network is normalized with the images of Mongolian Cyrillic character. Using back-propagation method efficient for programs with their own databases. LeNet-5 neural network regulate weight value and threshold value of the convolutional neural network and enable the program to achieve the minimum error. Finally, this article analyzes the training and testing error rate with research data.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Narankhuu Natsagdorj
PY  - 2016/11
DA  - 2016/11
TI  - Mongolian Handwriting Character Recognition Based On Convolutional Neural Network(CNN)
BT  - Proceedings of the 6th International Conference on Information Engineering for Mechanics and Materials
PB  - Atlantis Press
SP  - 580
EP  - 584
SN  - 2352-5401
UR  - https://doi.org/10.2991/icimm-16.2016.105
DO  - 10.2991/icimm-16.2016.105
ID  - Natsagdorj2016/11
ER  -