Probabilistic white strip approach to plastic bottle sorting system

One of the most important steps in plastic recycling industry is waste sorting. Plastic wastes are usually sort into two main categories, which are polyethylene terephthalate (PET) and non-PET. This paper proposes a probabilistic approach to automated plastic bottle sorting by integrating size, colour and distance modelling of the plastic waste. Firstly, white strips are identified by employing maximum likelihood approach. Information on the white and grey strips is then analyzed by using maximum a posteriori method. Feature histogram is built by factoring the output decision of each white strip with its size. Finally, likelihood test is performed to classify the waste into PET and non-PET. Our algorithm performs the best in all evaluation metrics compared to the benchmark algorithms. It is most suitable to be implemented in a factory with the ever changing surroundings.