Bachelor degree work was performed in collaboration with the ARH laboratory, under the supervision of Lucian Petrică.
Bachelor Thesis Description
The goal of the thesis work was the implementation of a video surveillance system utilizing a Parrot AR 2.0 drone to detect and track moving objects. The drone control was implemented utilizing the YADrone API in Java, providing the controls for rotation and lateral movement required for tracking, and video stream acquisition from the two cameras mounted on the drone. The video processing pipeline was implemented in OpenCV and consisted of background extraction, motion detection, object feature extraction, and feature tracking. Additional logic was implemented in Java to decide, based on the visual information available, whether the drone must move to maintain tracking. A graphical user interface was developed for debugging and demonstration