Non-destructive analysis of sucrose, caffeine and trigonelline on single green coffee beans by hyperspectral imaging

Abstract

Hyperspectral imaging (HSI) is a novel technology for the food sector that enables rapid non-contact analysis of food materials. HSI was applied for the first time to whole green coffee beans, at a single seed level, for quantitative prediction of sucrose, caffeine and trigonelline content. In addition, the intra-bean distribution of coffee constituents was analysed in Arabica and Robusta coffees on a large sample set from 12 countries, using a total of 260 samples. Individual green coffee beans were scanned by reflectance HSI (980–2500 nm) and then the concentration of sucrose, caffeine and trigonelline analysed with a reference method (HPLC-MS). Quantitative prediction models were subsequently built using Partial Least Squares (PLS) regression. Large variations in sucrose, caffeine and trigonelline were found between different species and origin, but also within beans from the same batch. It was shown that estimation of sucrose content is possible for screening purposes (R2 = 0.65; prediction error of ~ 0.7% w/w coffee, with observed range of ~ 6.5%), while the performance of the PLS model was better for caffeine and trigonelline prediction (R2 = 0.85 and R2 = 0.82, respectively; prediction errors of 0.2 and 0.1%, on a range of 2.3 and 1.1% w/w coffee, respectively). The prediction error is acceptable mainly for laboratory applications, with the potential application to breeding programmes and for screening purposes for the food industry. The spatial distribution of coffee constituents was also successfully visualised for single beans and this enabled mapping of the analytes across the bean structure at single pixel level.

Mapping the Spatiotemporal Distribution of Acid and Moisture in Food Structures during Gastric Juice Diffusion Using Hyperspectral Imaging

“…Calibration models generated using hyperspectral data with R 2 values of greater than 0.70 can be considered as good enough for prediction purposes. 13,28 Thus, it was observed that the PLS calibration models based on full spectra had a very good (R cal 2 > 0.70) 13,28 moisture prediction accuracy for all food matrices, with R cal 2 > 0.88 (Table 3 and Figure 2) and RMSEC < 0.48 ( Table 3). The predicted versus measured plots for the best PLS calibration models for moisture content are shown in Figure 2.…”

“…The resulting image spectra were processed to reduce scattering effects, using standard normal variate (SNV), the Savitzky-Golay smoothing process, and multiplicative scatter correction (MSC). 13,17 For each food slice image, the mean spectrum was extracted by averaging the spectra of each pixel in the region of interests. The mean spectral data for each sample were used in the next stage of the analysis.…”

Recognition of different Longjing fresh tea varieties using hyperspectral imaging technology and chemometrics

“…Image information can be gained while acquiring spectral information, so that the sample can be fully analyzed from a three‐dimensional data block (He et al, ; Wang, Sun, Pu, & Zhu, ). This technology has been widely used in agriculture and forestry research (Caporaso, Whitworth, Grebby, & Fisk, ; Zhang, Dai, & Cheng, ; Zhang, Feng, Liu, & He, ). It also has been used for tea identification.…”

“…This technology has been widely used in agriculture and forestry research (Caporaso, Whitworth, Grebby, & Fisk, 2017;Zhang, Dai, & Cheng, 2019;Zhang, Feng, Liu, & He, 2018). It also has been used for tea identification.…”

Determination of pH and acidity in green coffee using near‐infrared spectroscopy and multivariate regression

“…NIR spectroscopy is characterized by speed, reliability, simplicity, security, low operation cost, minimum sample processing, low instrumental maintenance, allowing various analyses simultaneously and not requiring the use of chemicals. 8,17,18 This technique has already proven useful for the prediction of several properties in coffee samples and has been applied in the determination of antioxidants, 19,20 caffeine, [21][22][23] cafestol and kahweol, 24 roasting degree prediction, 25 chemical composition of defective beans, 11 sensory characteristics 18,26,27 and geographical origin. 28 In this context, the objective of the work presented here was to evaluate the potential of NIR spectroscopy and spectral pre-processing methods associated with partial least squares regression (PLS-R) in determining the pH and acidity of green coffee samples.…”