Deep Sky Monitoring System for Mini-drone Detection and Tracking

The impacts of attacks based on Commercial-Off-The-Shelf (COTS) mini-UAVs, which have become more technologically advanced and cheaper, can be very dangerous with devastating effects. Therefore, countermeasures to monitor a no-fly zone are highly required. The aim of this work is to develop, design and release an AI powered continuous airspace monitoring system that defends a No-Fly Zone or a restricted area from ‘nuisance’ and ‘nefarious’ mini-drones. We have developed a sky deep-surveillance system using a multi-sensor camera and a convolutional neural network models, to perform near real-time analysis. Moreover, we have deployed the proposed sky monitoring system in a no-fly zone and evaluated the performance with which it could detect flying objects at different distances under different weather conditions. The experimental results prove the performance of the proposed continuous monitoring system and data analysis technique compared to other object detectors.