Construction Site Safety Image Dataset
Dataset composed of more than 5,600 images illustrating situations on construction sites, annotated to detect the wearing of personal protective equipment (helmets, vests, etc.), useful for the prevention of accidents and safety at work.
5,641 annotated images, JPEG/PNG formats, 14 annotation columns including PPE types and positions
CC BY 4.0
Description
The Construction Site Safety Image Dataset contains more than 5,600 annotated images, representing workers on construction sites with various personal protective equipment (PPE). Annotations detail the presence and position of helmets, vests, and other essential PPE.
What is this dataset for?
- Train computer vision models to detect the wearing of PPE in real time.
- Monitor compliance with safety standards on construction sites.
- Prevent accidents through better safety management.
Can it be enriched or improved?
Yes, it is possible to add additional annotations (e.g. type of PPE, compliance status), increase the volume with recent images or images specific to other environments, and create videos for dynamic detection.
🔎 In summary
🧠 Recommended for
- Machine vision researchers
- Safety engineers
- AI developers in construction
🔧 Compatible tools
- TensorFlow
- PyTorch
- Roboflow
- OpenCV
- LabelImg
💡 Tip
Use image augmentation techniques to improve robustness in a variety of conditions.
Frequently Asked Questions
What annotations are available in this dataset?
The dataset contains 14 annotation columns detailing the types of PPE and their positions on the images.
Can this dataset be used for real-time applications?
Yes, it is suitable for training detection models in real time on video or camera feeds.
Does the CC BY 4.0 license allow commercial use?
Yes, the license allows commercial use as long as the original author is mentioned.
