Award Date

May 2016

Degree Type

Thesis

Degree Name

Master of Science in Computer Science

Department

Computer Science

First Committee Member

Yoohwan Kim

Second Committee Member

Ajoy Datta

Third Committee Member

Venkatesan Muthukumar

Fourth Committee Member

Jisoo Yang

Number of Pages

72

Abstract

Unmanned Aerial Systems (UAS) are being used commonly for surveillance, providing valuable video data and reducing risk for humans wherever applicable. The cost of small UAS can range from as low as $30 to high as $5000, which makes it affordable by everyone. Most of the UAS are equipped with a camera which results in activities like disruption of privacy or capturing sensitive data. This research is aimed at developing a system which can detect and identify a drone and apply some counter measures to stop its functions or make it go away. The air traffic will increase significantly in next 20 years according to FAA (Federal Aviation Administration) and is doubling every year which increases the happening of all risks due to them. This system will detect and identify a drone through various methods combined into a single algorithm which includes image processing and audio detection. This research also proposes a new method where two algorithms run in sequence namely motion detection and SURF (Speed up Robust features). This system as a whole will warn a person when a drone is nearby and apply all the above mentioned techniques to keep the privacy intact and will also help keep the malicious or harmful drones away from the restricted/residential areas in the coming years.

Keywords

drone detection; Drone Tracking; Motion Detection; Speed up robust features

Disciplines

Computer Sciences

Language

English


Share

COinS