LCSD: Low-light Crack Segmentation Dataset

[ICRA 2024] CrackNex: a Few-shot Low-light Crack Segmentation Model Based on Retinex Theory for UAV Inspections.

Related Resources

• Code repository: https://github.com/zyaocoder/CrackNex
• Paper (arXiv): https://arxiv.org/abs/2403.03063

Dataset Overview

• 102 images well-illuminated crack images as training set
• 41 images 41 low-light crack images as test set

Download

Use command to download the dataset directly into your path:

huggingface-cli download chrisathy/LCSD --repo-type dataset --local-dir ./LCSD

Citation

If you use this dataset, please cite:

@inproceedings{yao2024cracknex,
  title={Cracknex: a few-shot low-light crack segmentation model based on retinex theory for uav inspections},
  author={Yao, Zhen and Xu, Jiawei and Hou, Shuhang and Chuah, Mooi Choo},
  booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)},
  pages={11155--11162},
  year={2024},
  organization={IEEE}
}

---