Litcius/Paper detail

Near-infrared hyperspectral circular polarization imaging and object classification with machine learning

Masato Suzuki, Kimitaka Doi, Moritsugu Sakamoto, Kohei Noda, Tomoyuki Sasaki, Nobuhiro Kawatsuki, Hiroshi Ono

2024Optics Letters11 citationsDOI

Abstract

We constructed a hyperspectral circular polarization ( S 3 ) imaging system in the near-infrared (NIR) region comprising a circularly polarized broadband light source, a polarization grating, and a commercial hyperspectral camera. With this system, we captured hyperspectral S 3 images of plastic samples. We then demonstrated the classification with machine learning and found that the hyperspectral S 3 images showed higher classification precision than the conventional NIR hyperspectral images. This result indicates that the hyperspectral S 3 imaging has potential for object classification even for samples with similar absorption spectra. This hyperspectral S 3 imaging system can be applied in garbage classification in recycling plants.

Topics & Concepts

Hyperspectral imagingFull spectral imagingNear-infrared spectroscopyArtificial intelligenceChemical imagingRemote sensingVNIROpticsComputer visionComputer sciencePhysicsGeologyOptical Polarization and EllipsometryRemote-Sensing Image ClassificationSpectroscopy Techniques in Biomedical and Chemical Research
Near-infrared hyperspectral circular polarization imaging and object classification with machine learning | Litcius