Inspiration
When I was working on an organic farm in Cambodia, I learned that farmers indiscriminately apply pesticides to protect their crops. While effective in the short-term, this erodes soil quality and destroys long-term harvest sustainability, not to mention having extreme negative health impact for consumers.
What it does
So we built IRL Debugger, which detects pests in real-time using object detection on video and images. It's meant to hook up to high-res cameras that will monitor crops, then send pesticide type and application dose to a smart fluid spray router.
How we built it
Roboflow, OpenAI models, Twitch/Chrome extensions, Firebase functions
Challenges we ran into
I don't want to talk about it 😒 (see link) link
Accomplishments that we're proud of
Identifying an actually extremely expensive and negative climate impact problem.
What we learned
Twitch extension development is a nightmare.
What's next for IRL Debugger
Not sure! But having more cameras, better labelling data, etc will increase accuracy.
Log in or sign up for Devpost to join the conversation.