On-line analysis for the classification of metal and plastic scrap using Laser-induced breakdown spectroscopy and machine learning algorithms
Several approaches to machine learning algorithms for the analysis of LIBS data collected from scrap pieces on conveyor belt have been tested. The methods include decision tree, artificial neural networks, and principal component analysis based techniques.
The performance of each algorithm in terms of classification accuracy and robustness is presented. Methods for data-preprocessing have also been tested which includes the normalization and selection of spectral features.
Autor(en)
Jonas Petersson, Baptiste Ottino, Arne Bengtson, Bertrand Noharet, Tania Irebo Schwartz
Contentpartner
© Institut für anthropogene Stoffkreisläufe
Publikation
Sensor-Based Sorting & Control - 2018 (März 2018)
Verkaufspreis
€ 5,00 oder im Premium-Abonnement enthalten