The Bin Buddy

Inspiration

Our team wanted to create a program that would solve an environmental problem we relate to. None of us can tell what goes in which waste collection, and realized that incorrect waste sorting in abundance harms the environment. Hence, The Bin Buddy was born!

What it does

This web application uses image recognition to help the citizens of Ontario sort their waste correctly.

How we built it

We used HTML to create the structure of the web app, CSS to design it, Python and Flask to code the back-end system, TensorFlow to develop the NLP model for the image recognition feature, and Keras to implement the neural network (done with Python).

Challenges we ran into

Not being able to switch between the links for the index and result files was a big issue for us. Also, creating the model was a very new and complex process for us.

Accomplishments that we're proud of

Learning and creating an ML project within 24 hours!! This is a huge accomplishment for the team!

What we learned

How to create an ML model, how to use Flask, and how to create an image recognition system.

What's next for The Bin Buddy

The next step for The Bin Buddy is for us to create a mobile friendly interface so that users can directly take pictures from their phones instead of uploading an image.

Built With

css3
flask
html5
jupyter
jupyter-notebook
keras
numpy
python
tensorflow

Submitted to

TerraHacks

Created by

I worked on developing the machine learning model using Tensorflow and trained it for accurate image classification on Jupyter Notebook. After training the model I saved it as a .kesar model file and loaded it to the flask web app and used its image classification to sort images that clients upload to labels of garbage, recycling, or compost.

Labib Sarwar
I worked on the front-end and gathering all the images so that the ml model can be trained. I also made the button that shows the output of what bin to put the material in.

Wesil Said
I worked on gathering most of the images that were used to train the machine learning model. I assisted with organizing that data. I was also in charge of the creative aspects of the project and helping my teammates with bugs.

neetid12 Dhiman
I worked on a bit of the back end and front end. I had to learn how to use the Flask AppBuilder to form our web app. It was a little challenging at first, however, I was able to understand and get the hang of it . Additionally, I assisted in organizing the data that was collected by my teammates.

sxdi1

Updates

neetid12 Dhiman started this project — Aug 04, 2024 06:59 AM EDT

Leave feedback in the comments!

Log in or sign up for Devpost to join the conversation.