EAD 2019（内窥镜伪影检测）数据集专注于内窥镜视频中伪影的检测与分割，是一个综合性的多模态数据集。该数据集汇集了来自6个不同数据中心的视频，涵盖胃镜、膀胱镜、胃食管和结肠镜等多种组织类型，以及白光、荧光和窄带成像等多种成像模式。数据集包含约2000帧训练视频和500帧测试视频，涉及像素饱和、运动模糊等多种伪影类型。为提高算法的泛化能力，数据集特别设计了约200帧来自不同数据中心的泛化挑战赛视频集。所有视频帧均由专家精心标注，包括边界框和语义分割掩膜，旨在促进伪影的精确识别与分割，提升内窥镜图像分析的质量和准确性。

内窥镜

Dataset
│
├── trainingData_detection
│   ├── 00000.jpg
│   ├── 00000.txt
│   └── ...
├── trainingData_semanticSegmentation
│   ├── 0_original_images
│       ├──00000.jpg
│       ├──00000_batch2.jpg
│       └── ...
│   ├── 00000.tif
│   ├── 00000_batch2.tif
│   └── ...

@article{SREP2020:ali,
title ={An objective comparison of detection and segmentation algorithms for artefacts in clinical endoscopy.},
author= {Ali, S., Zhou, F., Braden, B. et al.},
journal={Scientific Reports},
year={2020},
volume={10},
pages={2748},
doi={https://doi.org/10.1038/s41598-020-59413-5}
}

@misc{EAD2019endoscopyDatasetI,   
title={Endoscopy artifact detection (EAD 2019) challenge dataset},
author={Sharib Ali and Felix Zhou and Christian Daul and Barbara Braden and Adam Bailey and Stefano Realdon and James East and Georges Wagnières and Victor Loschenov and Enrico Grisan and Walter Blondel and Jens Rittscher},
year={2019},
eprint={1905.03209},
archivePrefix={arXiv},
 primaryClass={cs.CV}
}

@misc{EAD2019endoscopyDatasetII,
    title={A deep learning framework for quality assessment and restoration in video endoscopy},
    author={Sharib Ali and Felix Zhou and Adam Bailey and Barbara Braden and James East and Xin Lu and Jens Rittscher},
    year={2019},
    eprint={1904.07073},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

官方网站：
访问官网

下载链接：

相关论文：
查看论文

发布日期： 2019-08