Trash sorter

Trash Sorter is an AI app built solo. Using a pretrained neural network, it instantly classifies waste to guide proper disposal and boost recycling efficiency.

Comment

Inspiration

Recycling is easy to mess up. Contaminated bins, misclassified waste, and unclear guidelines make it hard for people to do the right thing. I wanted to create a tool that makes sorting trash simple, fast, and accurate, so small actions can have a bigger environmental impact. ♻️

What it does

Trash Sorter lets users upload an image of any waste item. Using a pre-trained neural network, it identifies the type of trash—Paper, Plastic, Cardboard, Metal, or Glass—and gives instant feedback. Users no longer need to guess; recycling becomes smart and effortless. 🧠🚮

How we built it

I developed the project solo, building the web interface and integrating the AI model. The core is a neural network classifier trained on labeled waste images, wrapped in a responsive, user-friendly web app that runs in the browser.

Challenges we ran into

  • Optimizing the web interface for fast image uploads and instant feedback.
  • Making a small-sample model generalize well to different trash types.

Accomplishments that we're proud of

  • Built an end-to-end AI-powered web app solo in a short time.
  • Achieved accurate classification across five trash categories.
  • Created a tool that is practical, educational, and environmentally conscious. 🌱

What we learned

  • How to integrate AI models into a functional web app.
  • The importance of data quality and real-world testing for neural networks.
  • Balancing speed, usability, and accuracy in a small hackathon project.

What's next for Trash Sorter

  • Expand the model to more trash categories.
  • Add real-time camera input for instant sorting without uploads.
  • Explore mobile app integration to make smart recycling accessible anywhere.
  • Build a real bot
Share this project:

Updates