The dataset used in the paper "Real-Time Detection of Personal Protective Equipment Violations for Construction Workers Using Semisupervised Learning and Video Clips", named CPPED (Construction Personal Protective Equipment Dataset), comprises a total of 2,612 images across 13 categories.
The CPPED data set created by this study can be downloaded at https://drive.google.com/drive/folders/1pd2_kzUA82M25s8HVL6rHCUUWCaTdVnv?usp=sharing .To obtain access to the CPPED dataset, you must submit a justified request detailing the intended use of the data. Please ensure that your request adheres to the guidelines and requirements specified in the dataset's license agreement.
Note:
- The dataset is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
- The CPPED Dataset includes images extracted from the research by Ferdous (2022) and the internet.
- Dataset annotation format: <
object-class> <x_center> <y_center> <width> <height>
Details of CPPEDv1:
| category | object num |
|---|---|
| head | 957 |
| person | 5936 |
| no gloves | 2017 |
| shoes | 1484 |
| glass | 701 |
| yellow | 1600 |
| white | 1563 |
| no shoes | 347 |
| blue | 604 |
| vest | 2587 |
| gloves | 985 |
| red | 862 |
| mask | 529 |
If you use our work, please consider citing:
@article{chen2025real,
title={Real-Time Detection of Personal Protective Equipment Violations for Construction Workers Using Semisupervised Learning and Video Clips},
author={Chen, Qihua and Long, Danbing and Wang, Siqi and Chen, Qirong and Yuan, Beifei},
journal={Journal of Construction Engineering and Management},
volume={151},
number={3},
pages={04024213},
year={2025},
publisher={American Society of Civil Engineers}
}