A Low-Cost System for Moisture Content Detection of Bagasse upon a Conveyor Belt with Multispectral Image and Various Machine Learning Methods

This research aimed to propose an online system based on multispectral images for the real-time estimation of moisture content (MC) sugarcane bagasse. The consisted a conveyor belt, four halogen bulbs, and camera. MC models were developed using machine learning algorithms, i.e., multiple linear regression (MLR), principal component (PCR), artificial neural network (ANN), PCA-ANN, Gaussian process (GPR), PCA-GPR, random forest (RFR), PCA-GPR. 150 samples (calibration set) meanwhile remaining 50 applied as validation set. comparison all showed that PCA-RFR model achieved better detection with higher accuracy prediction. best results which coefficient determination prediction (r2) 0.72, root mean square error (RMSEP) 11.82 wt%, ratio standard deviation (RPD) 1.85. show this technique was very useful rapid screening