Related Experiment Video
Updated: Jun 7, 2025
07:30
A Test Bed to Examine Helmet Fit and Retention and Biomechanical Measures of Head and Neck Injury in Simulated Impact
Published on: September 21, 2017
8.9K
High-Precision and Lightweight Model for Rapid Safety Helmet Detection
Xuejun Jia1,2, Xiaoxiong Zhou1, Chunyi Su1
1College of Electrical Engineering and Control Science, Nanjing Tech University, Nanjing 211816, China.
Sensors (Basel, Switzerland)
|November 9, 2024
Summary
This study enhances safety helmet detection in industrial settings by optimizing the YOLOv5s model. The improved model offers higher accuracy and computational efficiency for practical safety applications.
Area of Science:
- Computer Vision
- Machine Learning
- Industrial Safety
Background:
- Accurate safety helmet detection is crucial for industrial workplace safety.
- Existing models like YOLOv5s face challenges in accuracy and computational efficiency in complex industrial environments.
Purpose of the Study:
- To significantly improve the accuracy and computational efficiency of safety helmet detection.
- To optimize the YOLOv5s model for industrial applications.
Main Methods:
- Integrated the Convolutional Block Attention Module (CBAM) to enhance feature sensitivity.
- Implemented the Modified Penalty-Decay Intersection over Union (MPDIoU) loss function for improved bounding box regression.
- Adopted a lightweight MobileNetV3 architecture and replaced SE attention with CBAM to reduce model complexity.
Main Results:
- Reduced model parameters from 15.7 GFLOPs to 5.7 GFLOPs.
- Increased mean average precision (mAP) from 82.34% to 91.56%.
- Achieved superior performance in accuracy and computational efficiency.
Conclusions:
- The optimized YOLOv5s model demonstrates significant improvements for safety helmet detection.
- The enhanced model offers practical value for industrial safety monitoring systems.
