Nondestructive visualization and quantification of total acid and reducing sugar contents in fermented grains by combining spectral and color data through hyperspectral imaging

Total acid content (TAC) and reducing sugar content (RSC) are important evaluation indicators for the quality of fermented grains. In this study, the TAC and RSC of fermented grains were quantified using hyperspectral imaging (HSI). Two combined algorithms were used to extract the characteristic wavelengths of TAC and RSC. Nine color features of fermented grains were extracted based on H, S and V color channels. Multivariate analytical models were developed to predict TAC and RSC using full wavelengths, characteristic wavelengths, color features and fused data, respectively. The CF model established based on characteristic wavelengths extracted by CARS-SPA showed the best results in predicting TAC. Meanwhile, the PSO-SVR model built using fused data was the best model for predicting RSC. The visualization of the TAC and RSC was achieved using the optimal models. These results show that HSI can achieve non-destructive detection and visualization of TAC and RSC in fermented grains.