Application Research on Convolution Neural Network for Bridge Crack Detection
- 10.2991/iccia-17.2017.24How to use a DOI?
- image processing, bridge crack, detection, convolution neural network
The bridge crack detection still relies on human visual measurement nowadays, which means low efficiency and high cost. In view to this situation, the convolution neural network(CNN) was introduced into bridge crack detection to improve the efficiency and reduce the error caused by manual work. In this paper, we designed a recognition algorithm based on convolution neural network, which can directly input the crack images obtained by UAV, avoids the complicated feature extraction pre-processing used in the traditional image processing. Basis on that, this paper puts forward a algorithm of image filter window with variable size control, and also carries on the experiments to set the threshold value of the length of the connected domain. Finally, an intelligent recognition scheme is obtained, which is suitable for crack recognition, and the recognition rate is over 95%, means that it has a good application prospect.
- © 2017, 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/).
Cite this article
TY - CONF AU - Jinghang Cen AU - Jiankang Zhao AU - Xuan Xia AU - Chuanqi Liu PY - 2016/07 DA - 2016/07 TI - Application Research on Convolution Neural Network for Bridge Crack Detection BT - Proceedings of the 2nd International Conference on Computer Engineering, Information Science & Application Technology (ICCIA 2017) PB - Atlantis Press SP - 150 EP - 156 SN - 2352-538X UR - https://doi.org/10.2991/iccia-17.2017.24 DO - 10.2991/iccia-17.2017.24 ID - Cen2016/07 ER -