The article presents the development of a hyperspectral imaging system for plastic classification based on thermal emission in the LWIR range, with and without the influence of polyoxymethylene (POM). The objectives of the study are to examine the predictive power of a model for classifying plastic materials and to explore the potential of the technique for industrial applications.

The study uses several main tools. A low-order scanning Fabry-Perot interferometer, that enables imaging from 8 to 15 µm without external illumination and a thermal camera a microbolometer sensor to record interferograms of the plastics’ emission spectra. The data is then processed using principal component analysis and k-nearest neighbors algorithm.

Part of the setup included a black coated stainless steel sample covered with Acktar’s ultra black coating, used as a high emissivity blackbody reference to calibrate the emission spectra. This IR coating ensured accurate separation of subtle thermal features across different plastic types.

The results show that the developed hyperspectral camera system can classify the plastic types with high accuracy, 90% micro precision, and 99% macro specificity, except for ABS. There was a successful distinction 12 out of 13 tested plastics. The study concludes that the developed product is a promising tool for classifying commercial plastics based on their emission spectra in the long-wave IR regime.

The super black material was essential for distinguishing small emission differences needed for precise classification while emphasizing the importance of stray light reduction during hyperspectral IR imaging. The camera’s capability to perform effectively even with ultra-black materials demonstrates a reliable approach for future recycling automation and industrial quality control, where black-coated surfaces and infrared emission are crucial.

N. Larsen, A. L. Jørgensen, M. L. Henriksen, V. Petrunin, J. Kjelstrup-Hansen, B. Jørgensen, and M. Hinge, “Classification of commercial plastics using hyperspectral long wave infrared emission imaging,” Spectrochimica Acta – Part A: Molecular and Biomolecular Spectroscopy, vol. 329, Article 125501, 2025, doi: 10.1016/j.saa.2024.125501.