Introduction

The Safety Gear Detection application demonstrates using AI detection to identify if employees are wearing both their hard-hats and their high-visibility vests.

Benchmarks

The application includes a metrics overlay showing the camera, display, and inference timing details.

The Safety Gear Detection application contains two different models running: a person detection model and a safety-gear detection model. Statistics for each model will be noted separately where applicable. Because there are two models running, overall inference FPS is impacted.

Reported FPS and Resolutions

Item	Frames per Second	Resolution
Display	60.00	1920x1080
Camera	30.00	1920x1080
Inference	18.00 (total)	320x320 (person detection and safety detection)

Inference Timing Details

Item	Timing, person detection (ms)	Timing, safety detection (ms)
Time to load the frame into the model	1.81	7.85
Running model inference on the frame	7.95	40.1
Decoding boxes	0.24	1.67

Testing

The model used in this application was trained with the following specific personal protection equipment:

Hard-hats - Pyramex Cap Style 4-Point Ratchet Suspension Bump Cap (yellow)
High-Visibility vests - Pioneer Tear-Away Traffic High Visibility Safety Vest

Downloads