International Journal of Advances in Engineering & Scientific Research
Print ISSN : 2349 –4824
Online ISSN : 2349 –3607
Frequency : Continuous
Current Issue : Volume 12 , Issue 1
2025
Mrs. K.M.Swarnadevi, Abinaya S, Gowsika D, Priya T
Mrs. K.M.Swarnadevi, Abinaya S, Gowsika D, Priya T, Department of Computer Science and Engineering, Kongunadu College of Engineering and Technology, Tamilnadu, India
DOI : https://doi.org/10.5281/zenodo.14924843 Page No : 66-77
Published Online : 2025-02-25
Download Full Article : PDF Check for Updates
Ensuring Personal Protective Equipment (PPE) compliance in underground mining environments is crucial for worker safety, particularly in extreme conditions such as those found in the DraaSfar mine, the deepest mine in Morocco. This study proposes a real-time PPE compliance monitoring system that leverages advanced computer vision techniques, integrating object detection with pose estimation to enhance accuracy and reliability. A unique dataset was collected and annotated from the DraaSfar mine, capturing the challenges posed by its harsh environmental conditions. The study employs the newly developed YOLO Pose v8 algorithm for pose estimation and compares various object detection models, including YOLO v8, v9, v10, Real-Time Detection Transformer (RT-DETR), and YOLO World, for PPE detection. By integrating pose estimation key points, the system effectively filters out false detections, ensuring precise PPE compliance verification. Experimental results demonstrate that combining pose estimation with state-of-the-art object detection significantly improves detection rates, particularly under varying lighting and spatial conditions. The findings highlight the potential of this approach in enhancing safety monitoring and enforcement in underground mining operations.
Keywords: PPE compliance, underground mining, pose estimation, YOLO, RT-DETR, real-time monitoring, computer vision
