An LCTF-Based Multispectral Imaging System for Estimation of Apple Fruit Firmness: Part Ii. Selection of Optimal Wavelengths and Development of Prediction Models

Firmness of apple fruit is an important quality attribute, which varies in the same lot of fruit due to inherentbiological variability, climatic condition, cultural practice, harvest time or maturity level, and postharvest handling andstorage. This article reports on prediction of apple fruit firmness by using modified Lorentzian distribution (MLD) parametersof light scattering images, which were acquired by a compact multispectral imaging system over wavelengths between 650and 1000 nm. Optimal wavelengths for firmness prediction were determined with the multi-linear regression (MLR) andcross-validation method. In addition, diffuse reflectance spectra between 550 and 1100 nm were acquired from Red Deliciousand Golden Delicious apples by using a visible/near-infrared spectrometer (Vis/NIRS) in interactance mode. Multispectralscattering predictions of fruit firmness were compared with those from Vis/NIRS. For Red Delicious, seven wavelengths (690,770, 790, 810, 920, 980, and 1000 nm) were optimal for multispectral scattering prediction of fruit firmness, whereas forGolden Delicious, eight wavelengths (650, 690, 740, 750, 820, 880, 910, and 990 nm) were needed. The prediction modelsestablished based on the optimal wavelengths gave good firmness predictions with a correlation coefficient (r) of 0.82 anda standard error of validation (SEV) of 6.64 N for Red Delicious apples, and r = 0.81 and SEV = 6.58 N for Golden Deliciousapples. These results were considerably better than those obtained with Vis/NIRS.