| Abstrakt: |
Unmanned Aerial Vehicles (UAVs), commonly known as drones, have witnessed a significant surge in popularity across various sectors. Issues related to abuse have been put forward by the fast-growing number of drones operating in the national airspace, including those used for commercial and recreational purposes. Autonomous drone detection systems present a likely way to find the solution to the problem of possible drone abuse, including drug smuggling and privacy violations. Because drones and other objects in the sky might seem similar, it can be challenging to identify drones. Furthermore, to achieve high accuracy, automated drone detection systems must be trained with enough data. Additionally, real-time detection is required, but this calls for highly configured hardware, such a GPU. Concerns about privacy invasion, security lapses, and airspace violations have also increased with this spike. By utilizing sophisticated object recognition model called YOLOv5 (You Only Look Once version 5), this study presents a novel approach to drone detection. With the utilization of webcam feed and artificial intelligence, YOLOv5 is employed to recognize and detect drones in real time. The project contributes to improved security and privacy management in the era of UAV proliferation by integrating YOLOv5 with laptop accurate warnings. [ABSTRACT FROM AUTHOR] |
