Formulation of an improved hyperspectral image processing algorithm for food quality monitoring

Abstract

Sarawak's sago flour industry suffers from food fraud using whitening chemicals. This study uses Vis-NIR hyperspectral imaging and machine learning to rapidly detect adulterants. PLSR and PCR models excel in detecting calcium carbonate and alloxan monohydrate, respectively. A novel ensemble-based, nonlinear model was developed to enhance prediction accuracy. This research underscores the potential of hyperspectral imaging and machine learning for sago flour quality control.