Comparing the performance of miniaturized near-infrared spectrometers in the evaluation of mango quality

Near-infrared (NIR) spectroscopy is an effective tool for the non-destructive determination of produce quality parameters. Herein we report an evaluation of the performance of four different commercially available portable NIR spectrometers for the non-destructive determination of selected quality parameters of Nam-Dok Mai mangoes. The miniaturized NIR spectrometers: SCiO (750–1070 nm), Linksquare (400–1050 nm), DLP NIRscan Nano (900–1700 nm), and Neospectra (1300–2500 nm) were used to acquire NIR spectra from mango samples. The spectroscopic data was used to develop calibration models for quality parameters of Nam-Dok Mai mangoes. The investigated mango quality parameters were dry matter (DM), total soluble solids (TSS), titratable acidity (TA), pH, and firmness. The model development was carried out using partial least squares regression (PLSR). The results indicate that NIR spectroscopic data obtained using the SCiO and Linksquare instruments resulted in calibration models with satisfactory figures of merit (Rc2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${R}_{c}^{2}$$\end{document} > 0.80) for DM, TSS, and pH. In addition, spectroscopic data acquired with the Linksquare instrument could also be used to make a satisfactory calibration model for TA. On the other hand, figures of merit for the calibration models for firmness based on the spectroscopic data acquired using the SCiO and Linksquare were not satisfactory. The figures of merit for calibration models for all the investigated mango quality parameters, based on spectroscopic data acquired with the DLP NIRscan Nano and Neospectra spectrometers, were not satisfactory. However, the figures of merit for the calibration models based on the spectroscopic data acquired with the Neospectra spectrometer were better than the figures of merit for calibration models based on the spectroscopic data acquired with the DLP NIRscan Nano instrument.