Deep learning for chemometric analysis of plastic spectral data from infrared and Raman databases

Increasing plastic recycling rates is key to addressing plastic pollution. New technologies such as chemometric analysis of spectral data have shown great promises in improving the plastic sorting efficiency to boost recycling rates. In this work, a novel deep learning architecture, PolymerSpectraDecisionNet (PSDN) was developed, consisting of convolutional neural networks, residual networks and inception networks in a decision tree structure. To better represent the conditions in the plastic recycling industry, the models were built to identify the most widely recycled polymers - polyethylene, polypropylene and polyethylene terephthalate from open -sourced infrared and Raman spectral dataset containing over 20 different polymers. PSDN performed better than end-to-end neural networks, obtaining an accuracy of 0.949 and 0.967 with the Raman and infrared datasets respectively. The use of deep learning can also distinguish between weathered and unaged polymer samples, with accuracies of 0.954 for high density polyethylene and 0.906 for polyethylene terephthalate.

Automated summary

Increasing plastic recycling rates is crucial for addressing plastic pollution, and new technologies like chemometric analysis of spectral data show great potential for improving plastic sorting efficiency. In this study, a novel deep learning architecture called PolymerSpectraDecisionNet (PSDN) was developed to accurately identify commonly recycled plastics from infrared and Raman spectral datasets. PSDN outperformed end-to-end neural networks and demonstrated the ability to distinguish between weathered and unaged polymer samples.