This study addresses the critical challenge of detecting unmanned aerial vehicles (UAVs) in aerial surveillance tasks using deep learning-based object detection methods. With the increasing use of UAVs across various domains, ensuring reliable identification of UAVs, both authorized and unauthorized, is essential for airspace management and security. The study proposes the use of the YOLO11 algorithm, an advanced real-time object detection model, to improve the accuracy and robustness of UAV detection systems. The approach involves detecting both fixed-wing and rotary-wing UAVs from aerial images captured by a UAV-mounted high-resolution camera. A custom dataset was developed using UAV flight missions and image augmentation techniques to reflect real-world conditions. The main contribution of the study lies in evaluating and demonstrating the potential of YOLO11 in enhancing UAV detection performance in diverse environments. This work offers insights into the practical application of deep learning for real-time aerial monitoring and airspace security.