This project contains the source code of our paper "Rethinking Lightweight Convolutional Neural Networks for Efficient and High-quality Pavement Crack Detection". In the paper, we propose CarNet which is a fast pavement crack detector which achieves excellent accuracy. We also propose three new pavement crack datasets, namely Sun520, Rain365, and BJN260, to facilitate related research in the community.
Clone repo and install requirements.txt in a Python=3.6 environment, including PyTorch=1.6.
git clone https://github.com/shiyanrubing/CarNet-V1.0 # clone
cd CarNet-V1.0
pip install -r requirements.txt # install
Download datasets from here
Configure the specific dataset in the cfg.py, and run crack_train.py in command line
python crack_train.py
Run test.py in command line
python test.py
@article{li2021fast,
title={Fast and Accurate Road Crack Detection Based on Adaptive Cost-Sensitive Loss Function},
author={Kai Li and Bo Wang and Yingjie Tian and Zhiquan Qi},
journal={IEEE Transactions on Cybernetics},
pages={1-12},
year={2021},
doi={10.1109/TCYB.2021.3103885}
}
@article{li2023fast,
title={{Rethinking Lightweight Convolutional Neural Networks for Efficient and High-quality Pavement Crack Detection}},
author={Kai Li, Jie Yang, Siwei Ma, Bo Wang, Shanshe Wang, Yingjie Tian, and Zhiquan Qi},
journal="arXiv preprint arXiv:2109.05707",
year={2023}
}