Intelligent Surveillance Tower for Detection of the Drone from the Other Aerial Objects Using Deep Learning

The increasing use of drones for both civilian and military purposes has led to a growing need for effective drone detection systems. Conventional radar-based systems have limitations, including high costs, complex installation procedures, and limited accuracy in detecting small and low-flying drones. To address these limitations, this paper presents a novel drone detection system-based on deep learning using a modified You Only Look Once (YOLO) model. The system has been designed and implemented on hardware platforms to provide real-time and efficient drone detection. The hardware design includes a pre-processing module, a detection module, a feature analysis module, a tracking module, and an alarm module. The unique direction detection feature, enabled by a servo motor, adds an important layer of functionality to the system's already impressive capabilities. The system also predicts the trajectory of detected drones and can be integrated into existing security systems. The proposed deep learning-based approach is specifically optimized for drone detection, incorporating advanced techniques such as parameter adjustment and architectural modifications to improve performance. The system is scalable and adaptable to evolving threats, providing a highly accurate and efficient solution for real-time drone detection. This work represents a significant step forward in the effort to protect the public from the dangers posed by unregulated drone usage and offers a cost-effective and reliable alternative to traditional radar-based systems.