Real-Time AI-Driven Risk Detection and Safety Assessment in Electrical Maintenance Through 2D/3D/360 Degrees Vision Systems

This paper presents the development and implementation of an AI-based system designed to prevent accidents caused by human errors and improper use of Personal Protective Equipment (PPE). Named the “Safe Behavior Project,” the system integrates 3D stereo vision and neural networks for image analysis within geofenced areas. The use of 3D cameras enables depth-aware analysis and accurate positioning of objects and personnel. Real-time alerts are generated based on risk conditions, and violations are displayed immediately on the user interface. This approach surpasses conventional 2D methods by embedding AI into a spatially aware monitoring system.