An intelligent and vision-based system for Baijiu brewing-sorghum discrimination

Classification and characterization on sorghums based on HS-GC-IMS combined with OPLS-DA and GA-PLS

“…OPLS-DA can well solve samples with larger within-group divergence and obtain delightful classification effect ( Bylesjo et al, 2006 ). In previous study, accuracy of the anti-aliased convolutional network with machine vision obtained 89.15 % for ten cultivars ( Ma et al, 2022 ). Bai et al (2020) obtained 96 % accuracy of the model with HSI and PLS-DA.…”

“…In a previous study Bai et al (2020) achieved the identification accuracy of 91 % for the adulterated sorghums through HSI, Ma et al (2022) focused on several sorghum cultivars with one adulterated ratio and obtained the number of misclassified sorghum kernels through machine vision system. Our accuracy rate of adulteration discrimination is slightly lower but the samples were much more complicated.…”

“… Guindo et al (2016) predicted pericarp thickness of sorghum by NIRS and partial least squares-discriminant analysis (PLS-DA). Ma et al (2022) detected ten brewing-sorghum cultivars at single kernel sample level through machine vision system and the anti-aliased convolutional network. However, many sorghum cultivars have similar color and shape, visual recognition is limited to evaluate samples' appearance feature.…”