Deep Learning Based Multi-channel Road Crack Detection Method

DOI

10.2991/978-94-6463-304-7_59How to use a DOI?

Keywords

Crack detection; Deep learning; Multichannel; Convolutional neural networks

Abstract

With the growth of our economy, the construction of road infrastructure in the country is also developing vigorously. Under prolonged use, cracks of varying degrees can appear on road pavements. The detection of cracks on road surfaces is also an important part of road maintenance as well as road safety and security. Typically, traditional manual road crack detection is time-consuming and labour-intensive, and the results can be inaccurate due to differences in individual evaluation criteria. With the increasing maturity of deep learning algorithms, deep learning models are also being used for road crack detection. However, the currently used deep learning-based road crack detection methods are all based on the extraction of colour RGB images, and the information of the images is not fully utilised and the accuracy of the extraction needs to be improved. Therefore, this paper proposes a multi-channel road crack detection method based on deep learning, which improves the road crack detection accuracy by using RGB images and its grey map four channels as model inputs. Comparing the extraction results of deep learning models using only RGB three channels and four channels on the publicly available dataset CRACK500, it is found that the multi-channel road crack detection method proposed in this paper achieves an accuracy of 86.69%, which is ~10% better than the traditional RGB three channel detection method in terms of accuracy. It also provides a new idea for the road crack detection method based on deep learning.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Shengcheng Yang
AU  - Pengchao Jing
AU  - Zhixiong Guo
AU  - Shengxi Wang
AU  - Lei Gong
AU  - Dongdong Mu
AU  - Xingjun Cai
PY  - 2023
DA  - 2023/12/04
TI  - Deep Learning Based Multi-channel Road Crack Detection Method
BT  - Proceedings of the 3rd International Conference on Digital Economy and Computer Application (DECA 2023)
PB  - Atlantis Press
SP  - 567
EP  - 573
SN  - 2589-4900
UR  - https://doi.org/10.2991/978-94-6463-304-7_59
DO  - 10.2991/978-94-6463-304-7_59
ID  - Yang2023
ER  -