Safety vest detection: Datasets & Models
Computer vision detection that identifies whether workers are wearing high-visibility safety vests in images and video from construction and infrastructure sites. Flags people with and without proper vests in real time.
Safety vest detection
This detection workflow locates people in a scene and classifies whether each individual is wearing a high-visibility safety vest, flagging compliant and non-compliant workers with bounding boxes. It accepts still images or video frames from site cameras, drones, or mobile devices, and models can be trained on a safety vest dataset to fit specific site conditions.
Try it out
Powered by general detection model
Run on custom image
Drop an image here or click to upload
Why automate safety vest detection
On construction and infrastructure sites, PPE compliance is typically checked by supervisors doing manual spot inspections, which are inconsistent, hard to scale across large sites, and impossible to run continuously. Automating safety vest detection with computer vision turns existing camera feeds into an always-on monitoring layer that catches non-compliance the moment it happens. This reduces the administrative burden on safety officers and creates an auditable record of PPE adherence at entry gates, active work zones, and near heavy machinery.
Key benefits
Teams get continuous, objective monitoring of vest compliance without adding headcount, plus alerts that can trigger interventions before an incident occurs. The workflow integrates with the RTSP cameras and edge hardware most sites already run, and scales from a single checkpoint to multiple simultaneous zones. Starting from an existing safety vest dataset and the RF-DETR architecture shortens the path from prototype to a deployable model tuned to your site conditions.
How to use it
- Upload an image: Drop any photo into the widget above to try safety vest detection for free, right on this page.
- Preview instant results: The model detects safety vest, person, hard hat, worker without vest within seconds and returns an annotated image with structured predictions, so you can immediately see how well it works. Adjust the detection classes and re-run until the results fit your scenario.
- Like what you see?: With a free Roboflow account, you can deploy this workflow with the Serverless Cloud API, customize it in your workspace, or train RF-DETR on your own images for production accuracy.
- Get started for free: No code or training data required to begin: upload, review the results, and make it part of your stack in a single click.
Recommended model for safety vest detection
RF-DETR is state of the art for real-time object detection, instance segmentation, and keypoint detection, with higher accuracy and lower latency than comparable models, and it's Apache 2.0 licensed. When you're ready to build a production model for safety vest detection, fine-tune it on your own labeled images, then deploy to a Serverless Cloud API endpoint.
Not sure where to start?
Describe your Safety Vest Detection project to Roboflow Agent.