Table 3. Recent studies on meat quality detection using hyperspectral imaging (HSI) technique

Category Measured attribute Analytical method Performance References
Chicken meat Texture ACO-BPANN and PCA-BPANN Correlation coefficient of 0.754 (Khulal et al., 2016)
Prawn TVB-N (freshness) PLSR, LS-SVM, and BP-NN Correlation coefficient of 0.9547 (Dai et al., 2016)
Beef Total viable count (TVC) of bacteria (freshness) PLS and LS-SVM Accuracy of 97.14% (Yang et al., 2017a)
Pork meat Protein and TVB-N content PLSR and LS-SVM Correlation coefficient of 0.861 (Yang et al., 2017b)
Fish Freshness PCA and BP-ANN Accuracy of 94.17% (Huang et al., 2017)
Pork muscles Intramuscular fat contents SVM, SG, SNV, MSC, and PLSR Correlation coefficient of 0.9635 (Ma et al., 2018)
Frozen pork Myofibrils cold structural deformation degrees PLSR and SPA Correlation coefficient of 0.896 (Cheng et al., 2018)
Lamb, beef, and pork Adulteration SVM and CNN Accuracy of 94.4% (Al-Sarayreh et al., 2018)
Beef Adulteration PLSR and SVM Accuracy of 95.31% (Ropodi et al., 2017)
Fish (grass carp) Textural changes (Warner-Bratzler shear force, hardness, gumminess and chewiness) PLSR Correlation coefficient of 0.7982-Correlation coefficient of 0.8774 (Ma et al., 2017)
Lamb meat Adulteration SPA and SG Correlation coefficient above 0.99 (Zheng et al., 2019)
Pork Intramuscular fat content MLR Correlation coefficient of 0.96 (Huang et al., 2017)
Pork longissimus dorsi muscles Moisture content (MC) PLSR Correlation coefficient of 0.9489 (Ma et al., 2017)
Grass carp (Ctenopharyngodon idella) Moisture content PLSR Correlation coefficient of 0.9416 (Qu et al., 2017)
Lamb muscle Discrimination PCA, LMS, MLP-SCG, SVM, SMO, and LR Accuracy of 96.67% (Sanz et al., 2016)
Beef Adulteration PLSR, SVM, ELM, CARS, and GA Correlation coefficient of 0.97 (Zhao et al., 2019)
ACO, ant colony optimization; PCA, principle component analysis; BPANN, back propagation artificial neural network; PLSR, partial least squares regression; LS-SVM, least square support vector machines; BP-NN, back propagation neural network; PLS, partial least squares; SG, savitzky golay; SNV, smoothing, standard normal variate; MSC, multiplicative scatter correction; SPA, successive projections algorithm; CNN, convolution neural networks; LMS, linear least mean squares; MLP-SCG, multilayer perceptron with scaled conjugate gradient; SVM, support vector machine; SMO, sequential minimal optimization; LR, logistic regression; ELM, extreme learning machine; CARS, and competitive adaptive reweighted sampling; GA, Genetic algorithm.