The work focused on collecting imaging measurements of various substances present within a landfill, which were analyzed in a controlled environment using a digital multispectral SpeCtroscoPe (SCP). This sensor is designed to perform spectrophotometric measurements with 12 LEDs that emit light radiation in the visible (VIS) and near-infrared (NIR) ranges. The signal is collected by a photodiode and converted into a current signal, which is acquired and used as a dataset. The dataset was processed and analyzed with advanced machine learning tools such as a Multi-Layer Perceptron model (MLP). The collected data were then processed with machine learning methods to create algorithms that allow for the automatic sorting of waste within a landfill.
Fina, F., Magsi, K.U., Balsi, M., Piotto, M., Contardi, S., Leccese, F. (2026). LED Multispectral Sensor in Recycling Garbage. In Conference Proceedings - 2026 IEEE International Workshop on Metrology for Industry 4.0 and IoT, MetroInd4.0 and IoT 2026 (pp.242-247). Institute of Electrical and Electronics Engineers Inc. [10.1109/MetroInd4.0IoT69397.2026.11653187].
LED Multispectral Sensor in Recycling Garbage
Fina F.Writing – Original Draft Preparation
;Leccese F.
Writing – Review & Editing
2026-01-01
Abstract
The work focused on collecting imaging measurements of various substances present within a landfill, which were analyzed in a controlled environment using a digital multispectral SpeCtroscoPe (SCP). This sensor is designed to perform spectrophotometric measurements with 12 LEDs that emit light radiation in the visible (VIS) and near-infrared (NIR) ranges. The signal is collected by a photodiode and converted into a current signal, which is acquired and used as a dataset. The dataset was processed and analyzed with advanced machine learning tools such as a Multi-Layer Perceptron model (MLP). The collected data were then processed with machine learning methods to create algorithms that allow for the automatic sorting of waste within a landfill.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
