Real Time Safety Measurement and Hazard Notifications Using YOLOV8 and Django Framework
Author : S Sindhu and B Asha
Abstract :
Ensuring safety in industrial environments, construction sites, and public infrastructures is a critical challenge due to increasing hazards and reliance on manual supervision. Conventional safety monitoring systems depend on continuous human observation through CCTV cameras, which often results in delayed responses, human error, and overlooked safety violations. To address these limitations, this project proposes a Real-Time Safety Measurement and Hazard Notification System using YOLOv8 and Django.
The proposed system utilizes the YOLOv8 deep learning–based object detection model to analyse live video streams and identify hazardous situations such as the absence of safety helmets, fire incidents, and unauthorized access to restricted areas. The Django framework is used as the backend to manage data processing, event logging, and alert generation. When a hazard is detected, the system immediately triggers notifications to concerned authorities, enabling rapid preventive action.
By integrating real-time computer vision with a scalable web framework, the system significantly reduces dependency on manual monitoring, improves detection accuracy, and minimizes response time. The proposed solution enhances overall safety management and provides an efficient, automated approach for accident prevention in safety-critical environments.
Keywords :
YOLOv8, Real-Time Safety Monitoring, Hazard Detection, Computer Vision, Deep Learning, Django Framework, Object Detection, Automated Surveillance, Industrial Safety.
