A Novel Analyzing Method to Coal Mine Image Restoration

DOI

10.2991/ap3er-15.2015.67How to use a DOI?

Keywords

Image restorations;K-fold Cross-Validation;BP neural network; Coal mine

Abstract

In order to solve the phenomena that harsh coal mine environment will lead to coal mine monitoring image degradation, a K-fold Cross-Validation image restoration algorithm BP neural network was proposed. Firstly, the images will be blurred by Gaussian white noise. Then, the blurred image and original image match “training pairs”. When the training error and validation error is equal, stop the network training, select the training error and test error are smaller as the optimal model. Finally, bring the blurred image to the restoration model and image processing. Experiment shows that the K-Fold Cross-Validation BP neural network model for image restoration of generalization performance and fitting precision both meet the requirements.

Copyright

© 2015, 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  - Changchun Shang
PY  - 2015/06
DA  - 2015/06
TI  - A Novel Analyzing Method to Coal Mine Image Restoration
BT  - Proceedings of the 2015 Asia-Pacific Energy Equipment Engineering Research Conference
PB  - Atlantis Press
SP  - 285
EP  - 288
SN  - 2352-5401
UR  - https://doi.org/10.2991/ap3er-15.2015.67
DO  - 10.2991/ap3er-15.2015.67
ID  - Shang2015/06
ER  -