DeepCrack is a benchmark dataset for infrastructure crack detection provided by the Computer Vision and Remote Sensing Laboratory of Wuhan University. Related research papers include... DeepCrack: A deep hierarchical feature learning architecture for crack segmentationIt aims to provide standardized and high-precision supervised learning data support for crack detection algorithm research. It can be directly used for training and evaluation of deep learning models such as U-Net, DeepLab, and SegNet, and is widely used in research directions such as structural health monitoring, road inspection, and building defect identification. This dataset contains RGB crack images and their corresponding pixel-level binary labeled masks. All labels are manually labeled pixel by pixel, making it suitable for supervised semantic segmentation training. The dataset has been divided into training and test sets according to a standard structure, and each image corresponds to a mask file with the same name.

Citation

Liu et al., DeepCrack: A deep hierarchical feature learning architecture for crack segmentationNeurocomputing, 2019.

@article{liu2019deepcrack,
title={DeepCrack: A Deep Hierarchical Feature Learning Architecture for Crack Segmentation},
author={Liu, Yahui and Yao, Jian and Lu, Xiaohu and Xie, Renping and Li, Li},
journal={Neurocomputing},
volume={338},
pages={139--153},
year={2019},
doi={10.1016/j.neucom.2019.01.036}
}