Unmanned aerial vehicles (UAVs) play a crucial role in contemporary military operations, requiring the development of effective detection techniques to improve situational awareness and security. Current methods for identifying UAVs in video feeds typically involve manual inspection or traditional computer vision techniques. However, these approaches are known to be time-consuming and less accurate. In this study, the focus is on tackling the issue of real-time UAV detection. A potential solution is presented, which revolves around utilizing the YOLOv8 deep learning model. By leveraging publicly accessible datasets, the YOLOv8 model has been trained to identify unmanned aerial vehicles in military environments accurately. The assessment of the model's performance showcases its efficacy in precisely detecting UAVs in real-time video footage. This study utilizes deep learning techniques to provide a comprehensive solution for improving military intelligence and security by automating the detection of unmanned aerial vehicles.