Ensuring the safety corridor of high-voltage transmission lines is critical in construction sites and roadways intersecting with power lines. This study proposes a safety corridor violation warning system based on AIoT technology, integrating the YOLOv8n model deployed on a Raspberry Pi 5, using ONVIF-compatible IP cameras, 4G connectivity, and solar power. The system is compatible with widely available off-the-shelf IP cameras, helping reduce costs and simplify deployment. Experimental results show that the system achieves detection accuracy ranging from 80% to over 90% for heavy machinery such as cranes, pile drivers, and excavators operating near transmission lines. It maintains a processing speed of 4.45 frames per second and operates reliably in outdoor environments. Notably, the system can run continuously 24/7 using solar energy and is designed to remain operational for up to three days without sunlight, ensuring high availability under all weather conditions. Compared to traditional single-point sensor solutions, the proposed system supports wide-area monitoring, real-time alerts, and fully autonomous operation independent of the electrical grid.