Inspiration

According to Waste Management, an American trash collecting company, the average recycling contamination rate — or disposal of trash or recyclables in the wrong recycling bin — is 25%, meaning 1 in 4 items thrown in a recycling bin isn’t recyclable. For plastic products, less than 10% is typically recycled, with 75% of plastic ending up in landfills. (Source: link). This is in turn contributes to global warming and the destruction of ecosystems. We hope that our website can help solve this problem and preserve the beauty of the natural world.

What it does

Our code takes an image that the user uploads and then runs it though a neural network which has been trained to classify trash. Once the trash is classified, the class, along with instructions for disposal, are displayed. The code can be used by the common person to sort their own trash. However, we hope that with some minor changes, it also can be used in recycling plants and landfills to scan and detect waste that has been disposed of incorrectly

What's next for trash.ai

We hope to make our neural network more versatile and even more accurate, by giving it a more diverse database and adding a convolutional layer to it. In addition, a mobile app version could be developed to increase its utility for the common person.