Autonomous Drone-Based Border Surveillance Using Real-Time Object Detection with Yolo

Border surveillance constitutes a significant global concern, especially in isolated and susceptible areas prone to illicit crossings and security risks. This research introduces an innovative autonomous drone surveillance system intended for real-time monitoring and swift threat identification in border regions. The system utilises a custom-trained YOLO (You Only Look Once) object detection model version 11, built with Roboflow, to process live video feeds onboard for the precise detection of individuals and vehicles. The incorporation of machine learning facilitates accurate recognition across diverse environmental conditions, while instantaneous alerts improve situational awareness and enable swift responses by security personnel. This document delineates the system architecture, encompassing hardware integration, model training, and deployment strategy, succeeded by experimental validation in a simulated border scenario. The suggested method illustrates the capacity for scalable, intelligent, and autonomous border surveillance, representing a notable enhancement in aerial monitoring capabilities.