What it does
Unmanned Aerial Vehicle that recognizes "faces" while it flys by itself.
How we built it
JavaScript is used to automate the Ar drone 2.0, and also to take a live video stream from the drone and upload to the computer. The video is then parsed and analyzed with the Matlab to recognize faces.
Challenges we ran into
ALOT! First we wanted to use ROS, and its built in vision sensing capability so we installed linux, then to use install ROS we needed to install all the prerequisites packages. That route didn't work, so we started all over again with Javascript to automate the drone and Matlab for parsing video, and image analysis.
Accomplishments that we're proud of
None of us had any experience with JavaScript, and Linux
What we learned
Never use Linux on a Windows device.
During this project, first, we learn an operation of Drone with ROS system in Linux. The auto control system with JavaScript to change velocity, altitude and rotation. Also, how to analyze a picture and video to recognize faces with Matlab
What's next for Face Recognition for Drone
To actually recognize all type of faces, and later specific faces with complex features with machine learning techniques. And perhaps even tell who's face it is with Facebook's library to use it for aerial surveillance.
Log in or sign up for Devpost to join the conversation.