Recycling-oriented characterization of PET waste stream by SWIR hyperspectral imaging and variable selection methods

The proposed study was carried out to develop a fast and efficient strategy for plastic waste sensor-based sorting in recycling plants, based on hyperspectral imaging (HSI), combined with variable selection methods, produce high-quality recycled polyethylene terephthalate (PET) flakes stream. Variable techniques were applied order identify limited number of spectral bands useful recognize the presence other materials, considered as contaminant, inside stream PET flakes, reducing processing time requested by online applications. Post-consumer samples acquired HSI device working short-wave infrared (SWIR) range (1000 - 2500 nm). As first step, hypercubes processed applying chemometric logics build partial least squares discriminant analysis (PLSDA) classification model using full investigated range, able contaminant classes. second two different methods then applied, i.e., interval PLSDA (I-PLSDA) importance projection (VIP) scores, classes evaluate best method, showing efficiency values close those obtained spectrum model. result achieved VIP score method an average value 0.98. results suggested that variables can represent powerful approach sorting-online, decreasing amount data be thus enabling faster recognition compared