Every year millions if not billions of tons of plastic bottles land up in landfills or oceans and its because people do not sort their plastic bottle waste in proper categories.
What it does
There are 2 parts to this - mobile app and the machine. The app lets you take the photograph of the object and it will send the photo to the database which call an API from clarifai and determine which category the bottle falls into and finally returns a result in form of text. The machine is made of a webcam, servo motors and arduino. After you place the bottle in the machine and press a button the webcam will take a picture of the bottle and the process is same as with the app but after the API returns a result the machine will perform a task based on the result- if the bottle falls in to recycle category a servo motor will rotate and put the bottle in recycle bin, if the bottle falls in to trash category another servo motor will rotate and put the bottle in trash bin if it falls in reuse the a led will light up.
How we built it
We uploaded a bunch of pictures to clarifai API and made a custom trained computer vision model which could determine if a bottle was reusable, recyclable or trash. For the app we used firebase, flutter and dart. Initially we planned on using raspberry pi for the machine but we had to use arduino as they were not available.
Challenges we ran into
It was really hard to work on the machine without a raspberry pi as the communication between arduino and API is very limited.
Accomplishments that we're proud of
What we learned
What's next for Rehash
In future we plan to use raspberry pi instead of arduino. Also we wish to use optical sensor which can detect the microorganisms inside a bottle and really differentiate between a reusable bottle and recyclable bottle.