A lot of times, we are confused about what's recyclable and what's not. We want to put things in the right place, but a single object in the wrong place can contaminate the entire recycling bin. We wanted to find an automated way that would be time and cost efficient, while ensuring that we help the environment.
What it does
The sorting system identifies the material of the object thrown in the trash and sorts it into the correct bin (trash or recycling).
How we built it
We use a Raspberry Pi connected to a USB webcam to capture images of trash thrown in a bin which falls on a platform inside. We then use the Google Cloud Vision API to identify the object and its materials. The Raspberry Pi programs a Servo motor to tilt the platform towards the right bin (trash or recycling) based on the identified material.
Challenges we ran into
Establishing a headless connection with the Raspberry Pi definitely was a challenge, we eventually got it to work successfully thanks to everyone's help and a lot of interesting coding experiences! We also wanted to include compost in addition to waste and recycling, however the API was not able to differentiate between food wrappers and food scraps, so we decided to just do recycling and waste.
Accomplishments that we're proud of
We're proud of using the Google Cloud Vision API and getting the hardware (especially the Pi) to work. We learned a lot about both hardware and software.
What we learned
We learned how to utilize the Google Cloud Vision API, and how to get the Pi's IP address and establish connection headlessly and connect to VNC Viewer.
What's next for EasyCycle
We aim to use IR sensors to detect chemical composition in order to get more accurate identification of objects. We also will include a compartment for compost for identifying items such as food.
Log in or sign up for Devpost to join the conversation.