Trashform

Trashform Logo

Inspiration

Global recycling crisis stems from the widespread lack of awareness and motivation among the population. This leads to poor waste disposal practices and further contributes towards environmental degradation and many other harmful impacts. However, many individuals remain uninvested because of the absence of personal benefits.

What it does

Trashform is an AI-powered application that revolutionizes waste management through rigorous image recognition algorithms. Our users can upload their own photographs of waste items or use images online, and our system will classify the material as recyclable, compostable, or general waste. Our system will also determine the environmental impact through carbon footprint analysis and calculate the respective environmental impacts. Users will also receive recommendations for nearby recycling centers and trade-in locations where they can recycle these materials.

How we built it

Our project integrates a modern tech stack combining React with Vite for frontend performance and Tailwind CSS for responsive design implementation. The interface enables image upload functionality that communicates with our backend infrastructure. We also developed a sophisticated machine learning model using Python, training it on a comprehensive Kaggle dataset of over 19,000 classified waste images to achieve precise material classification.

Challenges we ran into

Integration between our frontend and backend systems presented significant technical issues, particularly dependency conflicts and package compatibility issues. TensorFlow also had compatibility limitations with the latest version of Python and was unable to render certain parts of our model.

Accomplishments that we're proud of

Through iterative development and rigorous testing, we successfully created three model training approaches, each requiring approximately 1.5 hours of computational time. Our final machine learning model achieved 93.5% accuracy rate in waste classification.

What we learned

This project helped us think more critically and develop solutions that address incentive gaps rather than replicating existing technologies. While recycling apps exist, the integration of image recognition technology with carbon footprint calculation and environmental analysis represents an approach that other projects have yet to do.

What's next for Trashform

Moving forward, Trashform will focus on expanding our material recognition database and enhancing model accuracy through advanced learning algorithms. Future development will include working with recycling companies to see how we can contribute the money from recycling the products towards sustainable projects.