Rock Thin Section Image Classification Research from Shallow Network to Deep Neural Network

DOI

10.2991/icemc-16.2016.125How to use a DOI?

Keywords

Rock thin section image classification; Shallow network; BP neural network; Deep neural network; Convolutional neural network; Deep learning

Abstract

Image classification technology has made tremendous progress with the development from shallow network (neural network) to deep neural network, image classification based on deep neural network has become popular in the field of image classification technology. By introducing shallow network and deep neural network, and making classification for 30 rock thin section images and contrast according to connectivity of pores with examples of the BP neural network (hidden layer has 6 neuron nodes and 7 neuron nodes) from shallow network and the convolutional neural network from deep neural network, finally, the average error rate of classification for deep neural network is 0%, and the average error rate of classification for BP neural network are 24.666% and 19.334% respectively, which shows that rock thin section image classification based on deep neural network has higher efficiency and better classification result than rock thin section image classification based on shallow network.

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  - Rongfang Gao
AU  - Chunxu Ji
AU  - Xinjian Qiang
AU  - Guojian Cheng
AU  - Ye Liu
PY  - 2016/05
DA  - 2016/05
TI  - Rock Thin Section Image Classification Research from Shallow Network to Deep Neural Network
BT  - Proceedings of the 2016 International Conference on Education, Management and Computer Science
PB  - Atlantis Press
SP  - 620
EP  - 625
SN  - 1951-6851
UR  - https://doi.org/10.2991/icemc-16.2016.125
DO  - 10.2991/icemc-16.2016.125
ID  - Gao2016/05
ER  -