Unmanned aerial vehicles (UAVs) have become increasingly important across a wide range of industries due to their versatility and ease of deployment. This study focuses on the development of a system for real-time object tracking and following, utilizing the DJI Tello drone. The DJI Tello is a lightweight quadcopter equipped with a 5-megapixel camera capable of capturing 720p video at 30 frames per second. The system employs computer vision techniques, particularly Convolutional Neural Networks (CNNs), to detect and track moving objects in real time, while dynamically adjusting the drone’s flight path to maintain continuous visual contact. The core functionality involves designing and testing an algorithm that processes the video feed from the drone’s camera and transmits flight commands to the drone's controller. The control system is essential for maintaining a safe distance from the target while avoiding collisions with surrounding obstacles. Python was used to communicate with the drone over Wi-Fi, issuing commands for take-off, landing, movement, and flight maneuvers. The system was tested under real-world conditions, such as tracking moving vehicles or pedestrians. By integrating the capabilities of the DJI Tello drone, advanced computer vision algorithms, and a robust control system, the proposed solution demonstrates potential to enhance UAV applications in disaster management, emergency response, and search and rescue operations.