Potato hierarchical clustering and doneness degree determination by near-infrared (NIR) and attenuated total reflectance mid-infrared (ATR-MIR) spectroscopy

摘要: Near-infrared (NIR) and attenuated total reflectance mid-infrared (ATR-MIR) spectroscopy were used to identify potato varieties detect doneness degree. The of tubers can be successfully classified by hierarchical cluster analysis (HCA). partial least squares regression (PLSR) model exhibited good prediction result for the degree evaluation. Principal component first-derivative iteration algorithm (PCFIA) was introduced select feature variables instead using full wavelength spectra modelling. Based on two sets selected from NIR MIR regions, both NIR–PCFIA–HCA MIR–PCFIA–HCA showed higher performances clustering. Moreover, NIR–PCFIA–PLSR MIR–PCFIA–PLSR models effectively predict tuber degree, yielding RP as high 0.935 RMSEP low 0.503. It is concluded that PCFIA an effective approach variable selection, spectroscopic techniques are capable classifying determining