Surveillance Drone: Smart Surveillance, A Detailed Study on Drones with Object Recognition Technology

Authors- Prof. Dr. Mohini Vyavhare, Mr. Tanmay Madhukar Dange, Mr. Dipak Gajanan Bitne, Mr. Sumit Hawre, Mr. Tejas Gedam, Mr. Sahil Pusatkar

Abstract--This paper presents a study on how drones have evolved from simple flying cameras to smart surveillance tools. Earlier, drones could only record videos without understanding what they captured. Now, with AI and computer vision, drones can detect, recognize, and classify objects like people, vehicles, and other items in real time. This makes drone surveillance faster, smarter, and more reliable. In our project, we built a smart surveillance drone that uses an FPV camera to stream live video. The video is analyzed by AI models, including YOLO, known for quick and accurate object detection. We tested the system in different conditions—daytime, nighttime, and crowded areas—to measure its performance. The drone successfully detected moving and still objects, though its accuracy dropped slightly in low light. Battery life was another limitation, reducing the overall flight time. The results show that smart surveillance drones can greatly help in public safety, traffic monitoring, and emergency response. The paper also suggests future improvements like better night vision sensors and longer battery life to make the system even more effective.

DOI: /10.61463/ijset.vol.13.issue3.143