EcoKiosk - Competition Submission
Tagline: Recycle. Get rewarded. It's that simple.
What is the problem you are solving and why does it matter?
Students don't recycle because there's no incentive. At my school, I watched plastic bottles go straight into the trash every lunch period. It wasn't that students didn't care. It was that recycling gave them nothing back. No recognition, no reward, no reason to bother.
The numbers confirm this. Texas recycles only 23% of its waste, which is 12 percentage points below the national average. Nationwide, only about 5% of plastic actually gets recycled. The infrastructure exists, but the motivation doesn't.
This matters because waste is growing faster than our ability to manage it. Celina, Texas, where I live, is the fastest-growing city in America with 830 new residents arriving every month. More people means more waste. If recycling rates stay low while the population grows, landfills overflow and plastic ends up in oceans.
The problem isn't awareness. Students know recycling is good. The problem is that recycling feels like a chore with zero payoff.
What is your solution and how does it work?
EcoKiosk is an AI-powered recycling station that rewards students for recycling. The process takes 10 seconds:
- Student holds up a plastic bottle to the camera
- The AI verifies it's real recyclable material (PET, HDPE, or PP)
- Student types their name on the touchscreen
- Student instantly earns school reward points
The AI model runs on a Raspberry Pi 5 using YOLO for object detection. It identifies plastic types in real time with a confidence threshold of 75% to prevent false positives. The interface is built with Python and PyQt5, designed to be fast and require no explanation.
The kiosk turns recycling from a chore into an achievement. Students compete to earn points. They recycle because it benefits them, not because someone told them to.
What is your execution or business plan?
Phase 1: Prove it works Deploy the first kiosk at my high school. Measure recycling rates before and after. Collect data on usage, points awarded, and student feedback.
Phase 2: Open source the project Release the code, hardware specs, and 3D print files for free. Any school can build their own kiosk for under $150 in parts.
Phase 3: Partner with schools Work with school districts to integrate EcoKiosk with existing reward systems like Minga. Offer a kit version for schools that don't want to build from scratch.
Phase 4: Scale Target the 130,000+ K-12 schools in the U.S. If even 1% adopt EcoKiosk, that's 1,300 schools changing student recycling behavior.
Revenue model (optional future path):
- Sell pre-built kiosk kits to schools
- License premium features like leaderboards and analytics dashboards
- Partner with recycling companies for sponsorship
Who are your target users or market?
Primary users: High school and middle school students ages 12 to 18
Secondary users: School administrators and sustainability coordinators who want to increase recycling rates without adding staff workload
Market size:
- 130,000+ K-12 schools in the United States
- 50 million+ students
- Growing demand for sustainability initiatives in education
Why students? Students are habit-forming. If they learn to recycle in school because it rewards them, they carry that behavior home. Changing student behavior today changes adult behavior tomorrow.
Technologies Used
Python, PyQt5, OpenCV, YOLO, TensorFlow, Raspberry Pi 5, Pi Camera Module 3, GPIO, FreeCAD
My Contribution
I designed and built the entire project from the ground up. I identified the problem, researched the technology, and developed both the hardware and software systems.
- Problem Research: Analyzed recycling statistics for Texas and observed student behavior at my school
- Hardware Design: Selected and integrated Raspberry Pi 5, camera module, and touchscreen. Designed the enclosure in FreeCAD for 3D printing
- Software Development: Wrote the kiosk application in Python using PyQt5 and OpenCV
- AI Model: Implemented YOLO for real-time plastic classification and tuned the confidence threshold to minimize false positives
- Testing: Tested in real-world conditions, identified lighting and speed issues, and optimized for edge hardware
- Demo: Built a web-based simulator to demonstrate the kiosk experience
Log in or sign up for Devpost to join the conversation.