Due to the complexity and the diversity of microplastics, several methods can be found in the literature for the detection of these particles. This paper presents a novel detection model for microplastics particles based on binary discrimination between two signals: noisy signal and particle signal over noise. The problem is solved in probabilistic terms under recording corrupted by normal noisy. The model is tested over several synthetic signals, and the performances are evaluated varying both the signal-noise ratio and the dataset balance distribution (the a priori probability of measuring the particle signal P-1 within the set of recorded signal segments). The model can be easily implemented into a microcontroller forming an embedded system for real-time fluid monitoring.
Online detection of floating microplastics in liquids
Sabatini A.;Vollero L.
2022-01-01
Abstract
Due to the complexity and the diversity of microplastics, several methods can be found in the literature for the detection of these particles. This paper presents a novel detection model for microplastics particles based on binary discrimination between two signals: noisy signal and particle signal over noise. The problem is solved in probabilistic terms under recording corrupted by normal noisy. The model is tested over several synthetic signals, and the performances are evaluated varying both the signal-noise ratio and the dataset balance distribution (the a priori probability of measuring the particle signal P-1 within the set of recorded signal segments). The model can be easily implemented into a microcontroller forming an embedded system for real-time fluid monitoring.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
