Inspiration
Our inspiration was that we wanted to create a machine learning algorithm. Both AI and drones are a hot topic these days, so we thought it would be fun to do a project combining the two.
What it does
This project is a comprehensive overview of using deep-learning-based object detection methods for Lunar simulation field imagery via drones.
How we built it
We used Python OpenCV and pre-existing code that we modified to detect a specific item on the terrain.
Challenges we ran into
It was difficult using OpenCV since it was a new experience, and what was even more difficult was this new concept of machine learning and trying to wrap our heads around the algorithms that can be confusing to understand for the first time.
Accomplishments that we're proud of
We're very proud of the fact that we have a model that can train and test for polar bears on a terrain, and all of us learned so many new things related to neural networks.
What we learned
We explore some of these applications along with challenges in the automation of drone-based monitoring through deep learning.
What's next for Game of Drones - ASRC Federal
We want to improve our code so that it can classify more than two states. In other words, instead of determining whether a polar is or is not in the picture, we want to know what type of item is in the picture. We'd also like to detect images and track it on the screen, rather than only confirm its existence.
Log in or sign up for Devpost to join the conversation.