Quantitative Evaluation of Color, Firmness, and Soluble Solid Content of Korla Fragrant Pears via IRIV and LS-SVM

Customers pay significant attention to the organoleptic and physicochemical attributes of their food with improvement living standards. In this work, near infrared hyperspectral technology was used evaluate one-color parameter, a*, firmness, soluble solid content (SSC) Korla fragrant pears. Moreover, iteratively retaining informative variables (IRIV) least square support vector machine (LS-SVM) were applied together construct evaluating models for quality parameters. A set 200 samples chosen its data acquired by using a imaging system. Optimal spectral preprocessing methods selected obtain out partial regression (PLSRs). The results show that combination multiplicative scatter correction (MSC) Savitsky-Golay (S-G) is most effective method parameters fruit. Different characteristic wavelengths a* value, SSC pears, respectively, after 6 iterations. These values obtained via IRIV reverse elimination method. correlation coefficients validation measure 0.927, 0.948, 0.953, respectively. Furthermore, error weight, γ, kernel function σ2, same (8.67 × 104, 1.21 103), (1.45 2.93 104), (2.37 105, 3.80 This study demonstrates LS-SVM can be pears define grade.