Waste Classification Algorithm
share
Waste sorting is a major environmental problem. Many have a hard time determining whether waste is organic, meaning food or natural material. Versus recyclable, which are able to be processed and used again. Computer Vision may be used to combat this problem. Our idea is to build and utilize a Convolutional Neural Network (CNN) to classify recyclable and organic waste. We use the Pytorch Machine Learning Library in our CNN. The model uses a simple CNN model with binary cross entropy to classify through images from our dataset as recyclable (R) or organic (O). The model is about 80-90% successful in its final state. In the majority of cases, it can quickly and successfully determine if an object is recyclable or organic.
READ FULL TEXT
