Due to the exponential rise in population, waste has also increased exponentially. This increase in waste has been proven to be detrimental to the environment and cause problems such as pollution, global warming, etc. Since Donald Trump has now become the President of the United States and has chosen to disband the EPA, it is now our responsibility to keep the environment in check for future generations.
What it does
It is a system which guides people into maintaining a cleaner environment by scanning trash and classifying it as compost, recycle, or landfill.
How we built it
We created a tensor flow environment on raspberry pi which displayed the outputs of the neural network as a character. The program starts when the button is pressed on the breadboard which takes a picture and scans it and classifies the image as compost, recycle, or landfill against the neural network. According to the classification, the result is outputted as a text file and then displayed on the corresponding LED on a breadboard.
Challenges we ran into
The raspberry pi does not have the computational power to run the convolutional neural network that we trained for image classification. Because of this we had to create a contrived demo
Accomplishments that we're proud of
We were able to correctly classify the objects manually. We were able to implement a user-friendly process which ran the system by the press of a button.
What we learned
We learned how to implement tensor flow in raspberry pi.
What's next for WasteSort
Implement a system with more memory and create a faster design.