EdmCrack600: a Pixel-level annotated dataset for crack identification

The dataset consists of 600 images of pavement cracks taken from roads in Edmonton, Canada. They are all annotated at pixel level to test automatic crack detection algorithms.

The link to down load the dataset: https://drive.google.com/file/d/1TJ10eoUmd3N2SBBalzj3On9J3S4xJu6M/view?usp=sharing

Please cite our relevant papers below if you are going to use this dataset for your academic purpose. Commercial use is not allowed.

1. Q. Mei, M. Gül, and M.R. Azim, Densely connected deep neural network considering connectivity of pixels for automatic crack detection. Automation in Construction, 2020. 110: p. 103018. (PDF)
2. Q. Mei, and M. Gül, A cost effective solution for pavement crack inspection using cameras and deep neural networks. Construction and Building Materials, 2020. 256: p. 119397. (PDF)
3. Q. Mei, M. Gül, and N. Shirzad‐Ghaleroudkhani (2020) "Towards Smart Cities: Crowdsensing-based Monitoring of Transportation Infrastructure using Moving Vehicles." Journal of Civil Structural Health Monitoring. (PDF)