Integrated Microwave Radar and Camera for Object Identification

Abstract

The rapid proliferation of unmanned aerial vehicles (UAVs) in both civilian and military sectors has introduced substantial challenges to surveillance, defense, and public safety systems. Small-scale drones, in particular, are difficult to detect due to their low radar cross-sections, silent flight capability, and ability to operate at low altitudes. Conventional single- layer detection techniques based solely on radar, acoustic sensing, or vision-based approaches often fail to deliver reliable accuracy across varying operational environments. To address these limitations, this paper presents a dual-layer UAV detection and neutralization framework that integrates low-cost microwave radar with artificial intelligence (AI)- driven computer vision. The primary detection layer employs an microwave radar module, enabling continuous 360° motion detection and real-time tracking of aerial targets. The secondary layer leverages deep learning-based computer vision, specifically a YOLO (You Only Look Once) architecture, to classify and confirm UAVs while distinguishing them from non-threatening aerial objects. Upon classification of a potential threat, the system activates a laser-based neutralization mechanism and concurrently transmits alerts to security operators for situational awareness.