EcoSort

Identifies recycling items
Has location selection feature
Accurately identifies the item and the waste category it belongs to

Inspiration:

Our journey began with a simple, yet profound realization: sorting waste is confusing! We were motivated by the challenge many face in distinguishing recyclables from garbage, and we saw an opportunity to leverage technology to make a real environmental impact. We aimed to simplify recycling, making it accessible and accurate for everyone.

What it does:

EcoSort uses a trained ML model to identify and classify waste. Users present an item to their device's webcam, take a photo, and our website instantly advises whether it is recyclable or garbage. It's user-friendly, efficient, and encourages responsible waste disposal.

How we built it:

We used Teachable Machine to train our ML model, feeding it diverse data and tweaking values to ensure accuracy. Integrating the model with a webcam interface was critical, and we achieved this through careful coding and design, using web development technologies to create a seamless user experience.

Challenges we ran into:

The most significant challenge was developing a UI that was not only functional but also intuitive and visually appealing. Balancing these aspects took several iterations.
Another challenge we faced, was the integration of our ML model with our UI.
Ensuring our ML model accurately recognized a wide range of waste items was another hurdle, requiring extensive testing and data refinement.

Accomplishments that we're proud of:

What makes us stand out, is the flexibility of our project. We recognize that each region has its own set of waste disposal guidelines. To address this, we made our project such that the user can select their region to get the most accurate results. We're proud of creating a tool that simplifies waste sorting and encourages eco-friendly practices. The potential impact of our tool in promoting environmentally responsible behaviour is something we find particularly rewarding.

What we learned:

This project enhanced our skills in ML, UI/UX design, and web development. On a deeper level, we learned about the complexities of waste management and the potential of technology to drive sustainable change.

What's next for EcoSort:

We plan to expand our database to accommodate different types of waste and adapt to varied recycling policies across regions. This will make EcoSort a more universally applicable tool, further aiding our mission to streamline recycling for everyone.
We are also in the process of hosting the EcoSort website as our immediate next step. At the moment, EcoSort works perfectly fine locally. However, in regards to hosting the site, we have started to deploy it but are unfortunately running into some hosting errors.
Our site is currently working