Review of the Development of an AI-Enabled Powerline Security System Using Infrared Image Processing and a Fuzzy Logic Threat Classification for Real-Time Intrusion Detection (Published)
This paper introduces an all-in-one, smart, surveillance system to protect high-voltage transmission lines with a combination of infrared (IR) thermal imaging, artificial intelligence (AI)-based object detection, fuzzy logic-based threat classification, geolocation tagging, and real-time wireless alerts. The framework solves the endemic problem of ensuring that remote and low visibility transmission routes are not intruded, vandalized and sabotaged- problems which traditional security mechanisms find hard to control. Using deep learning models like YOLOv8 and Convolutional Neural Network (CNNs), the suggested system is able to improve detection and situational awareness even in various environmental parameters. An interpretation fuzzy inference system also puts the detected events more into context by evaluating proximity, time at which events take place, and thermal intensity, which reduces false alarms. The paper is a critical review of the contemporary developments in AI-aided thermal surveillance, existing gaps in the existing methodologies, and the viability of implementing real-time, edge-enabled thermal surveillance systems on a large-scale power network. The study is a contribution to the emerging body of intelligent infrastructure protection since it indicates how AI-based systems can be used to change reactive security-focused systems into proactive ones.
Keywords: AI-enabled powerline security system, fuzzy logic threat classification, infrared image processing, real-time intrusion detection