Rapid and non-destructive prediction of mango quality attributes using Fourier transform near infrared spectroscopy and chemometrics

A decision support tool to analyze the properties of wheat, cocoa beans and mangoes from their NIR spectra

Near infrared spectroscopy (NIRS) is an analytical technique that offers a real advantage over laboratory analysis in the food industry due to its low operating costs, rapid analysis, and non-destructive sampling technique. Numerous studies have shown the relevance of NIR spectra analysis for assessing certain food properties with the right calibration. This makes it useful in quality control and in the continuous monitoring of food processing. However, the NIR calibration process is difficult and time-consuming. Analysis methods and techniques vary according to the configuration of the NIR instrument, the sample to be analyzed and the attribute that is to be predicted. This makes calibration a challenge for many manufacturers. This paper aims to provide a data-driven methodology for developing a decision support tool based on the smart selection of NIRS wavelength to assess various food properties. The decision support tool based on the methodology has been evaluated on samples of cocoa beans, grains of wheat and mangoes. Promising results were obtained for each of the selected models for the moisture and fat content of cocoa beans (R2cv: 0.90, R2test: 0.93, RMSEP: 0.354%; R2cv: 0.73, R2test: 0.79, RMSEP: 0.913%), acidity and vitamin C content of mangoes (R2cv: 0.93, R2test: 0.97, RMSEP: 17.40%; R2cv: 0.66, R2test: 0.46, RMSEP: 0.848%), and protein content of wheat-DS2 (R2cv: 0.90, R2test:0.92, RMSEP: 0.490%) respectively. Moreover, the proposed approach allows results to be obtained that are better than benchmarks for the moisture and protein content of wheat-DS1 (R2cv: 0.90, R2test: 94, RMSEP: 0.337%; R2cv: 0.99, R2test: 0.99, RMSEP: 0.177%), respectively. PRACTICAL APPLICATION: This research introduces a practical tool aimed at determining the quality of food by identifying specific light wavelengths. However, it is important to acknowledge potential challenges, such as overfitting. Before implementation, it is crucial for further research to address and mitigate the issues to ensure the reliability and accuracy of the solution. If successfully applied, this tool could significantly enhance the accuracy of near-infrared spectroscopy in assessing food quality attributes. This advancement would provide invaluable support for decision-making in industries involved in food production, ultimately leading to better overall product quality for consumers.

Recent advancements in NIR spectroscopy for assessing the quality and safety of horticultural products: A comprehensive review

The qualitative and quantitative evaluation of agricultural products has often been carried out using traditional, i.e., destructive, techniques. Due to their inherent disadvantages, non-destructive methods that use near-infrared spectroscopy (NIRS) coupled with chemometrics could be useful for evaluating various agricultural products. Advancements in computational power, machine learning, regression models, artificial neural networks (ANN), and other predictive tools have made their way into NIRS, improving its potential to be a feasible alternative to destructive measurements. Moreover, the incorporation of suitable preprocessing techniques and wavelength selection methods has arguably proven its practical feasibility. This review focuses on the various computation methods used for processing the spectral data collected and discusses the potential applications of NIRS for evaluating the quality and safety of agricultural products. The challenges associated with this technology are also discussed, as well as potential future perspectives. We conclude that NIRS is a potentially useful tool for the rapid assessment of the quality and safety of agricultural products.

Field Application of a Vis/NIR Hyperspectral Imaging System for Nondestructive Evaluation of Physicochemical Properties in ‘Madoka’ Peaches

Extensive research has been performed on the in-field nondestructive evaluation (NDE) of the physicochemical properties of ‘Madoka’ peaches, such as chromaticity (a*), soluble solids content (SSC), firmness, and titratable acidity (TA) content. To accomplish this, a snapshot-based hyperspectral imaging (HSI) approach for filed application was conducted in the visible and near-infrared (Vis/NIR) region. The hyperspectral images of ‘Madoka’ samples were captured and combined with commercial HSI analysis software, and then the physicochemical properties of the ‘Madoka’ samples were predicted. To verify the performance of the field-based HSI application, a lab-based HSI application was also conducted, and their coefficient of determination values (R²) were compared. Finally, pixel-based chemical images were produced to interpret the dynamic changes of the physicochemical properties in ‘Madoka’ peach. Consequently, the a* values and SSC content shows statistically significant R² values (0.84). On the other hand, the firmness and TA content shows relatively lower accuracy (R² = 0.6 to 0.7). Then, the resultant chemical images of the a* values and SSC content were created and could represent their different levels using grey scale gradation. This indicates that the HSI system with integrated HSI software used in this work has promising potential as an in-field NDE for analyzing the physicochemical properties in ‘Madoka’ peaches.

Robust prediction performance of inner quality attributes in intact cocoa beans using near infrared spectroscopy and multivariate analysis

Fast and simultaneous determination of inner quality parameters, such as fat and moisture contents, need to be predicted in cocoa products processing. This study aimed to employ the near-infrared reflectance spectroscopy (NIRS) in predicting the quality mentioned above parameters in intact cocoa beans. Near-infrared spectral data, in a wavelength ranging from 1000 to 2500 nm, were acquired for a total of 110 bulk cocoa bean samples. Actual fat and moisture contents were measured with standard laboratory procedures using the Soxhlet and Gravimetry methods, respectively. Two regression approaches, namely principal component regression (PCR) and partial least square regression (PLSR), were used to develop the prediction models. Furthermore, four different spectra correction methods, namely multiple scatter correction (MSC), de-trending (DT), standard normal variate (SNV), and orthogonal signal correction (OSC), were employed to enhance prediction accuracy and robustness. The results showed that PLSR was better than PCR for both quality parameters prediction. Spectra corrections improved prediction accuracy and robustness, while OSC was the best correction method for fat and moisture content prediction. The maximum correlation of determination (R²) and residual predictive deviation (RPD) index for fat content were 0.86 and 3.16, while for moisture content prediction, the R² coefficient and RPD index were 0.92 and 3.43, respectively. Therefore, NIRS combined with proper spectra correction method can be used to rapidly and simultaneously predict inner quality parameters of intact cocoa beans.

Prediction of chemical parameters and authentication of various cold pressed oils with fluorescence and mid-infrared spectroscopic methods

It was aimed to compare the performances of two spectroscopic methods, fluorescence and mid-infrared spectroscopy, in terms of their adulteration detection and estimation of several chemical properties for various cold pressed seed oils. Spectroscopic profiles, fatty acid, free fatty acid and total phenol contents of pumpkin seed, grape seed, black cumin oil, and sesame seed oils were determined and these oils were mixed with sunflower oil at 1-50% (v/v). Both spectroscopic techniques provided comparable results for determination of adulteration of each oil type and the most successful prediction was obtained for pumpkin seed oil at levels >%1. Combined data set of oils resulted in successful quantification of their free fatty acid value, total phenol and major fatty acids contents with both spectroscopic methods regardless of oil type. Both techniques could be used as reliable, fast and environmentally friendly alternatives in the analyses of different types of seed oils.

Reliable Discrimination of Green Coffee Beans Species: A Comparison of UV-Vis-Based Determination of Caffeine and Chlorogenic Acid with Non-Targeted Near-Infrared Spectroscopy

Species adulteration is a common problem in the coffee trade. Several attempts have been made to differentiate among species. However, finding an applicable methodology that would consider the various aspects of adulteration remains a challenge. This study investigated an ultraviolet–visible (UV-Vis) spectroscopy-based determination of caffeine and chlorogenic acid contents, as well as the applicability of non-targeted near-infrared (NIR) spectroscopy, to discriminate between green coffee beans of the Coffea arabica (Arabica) and Coffea canephora (Robusta) species from Java Island, Indonesia. The discrimination was conducted by measuring the caffeine and chlorogenic acid content in the beans using UV-Vis spectroscopy. The data related to both compounds was processed using linear discriminant analysis (LDA). Information about the diffuse reflectance (log 1/R) spectra of intact beans was determined by NIR spectroscopy and analyzed using multivariate analysis. UV-Vis spectroscopy attained an accuracy of 97% in comparison to NIR spectroscopy’s accuracy by selected wavelengths of LDA (95%). The study suggests that both methods are applicable to discriminate reliably among species.

Enhanced near infrared spectral data to improve prediction accuracy in determining quality parameters of intact mango

Presented manuscript aimed to describes enhanced near infrared spectral dataset used to improve prediction performances of near infrared models in determining quality parameters of intact mango fruits. The two mentioned quality parameters are total acidity (TA) and vitamin C which corresponds to main inner attributes of fruits. Near infrared (NIR) spectra data were acquired and recorded as absorbance spectral data in wavelength range from 1000 to 2500 nm. These data were then enhanced by means of several algorithms like multiplicative scatter correction (MSC), baseline linear correction (BLC) and combination of them (MSC+BLC). Prediction models, used to determine TA and vitamin C were established using most common approach: partial least square regression (PLS) based on raw and enhanced spectral data respectively. Prediction performances can be evaluated based on prediction accuracy and robustness, by looking statistical indicators presented as coefficient of determination (R²) and correlation (r), root mean square error (RMSE) and residual predictive deviation (RPD). Enhanced NIR spectral dataset can be employed as a rapid, effective and non-destructive method to determine inner quality parameters of intact fruits.

Rapid Prediction of Moisture Content in Intact Green Coffee Beans Using Near Infrared Spectroscopy

Moisture content (MC) is one of the most important quality parameters of green coffee beans. Therefore, its fast and reliable measurement is necessary. This study evaluated the feasibility of near infrared (NIR) spectroscopy and chemometrics for rapid and non-destructive prediction of MC in intact green coffee beans of both Coffea arabica (Arabica) and Coffea canephora (Robusta) species. Diffuse reflectance (log 1/R) spectra of intact beans were acquired using a bench top Fourier transform NIR instrument. MC was determined gravimetrically according to The International Organization for Standardization (ISO) 6673. Samples were split into subsets for calibration (n = 64) and independent validation (n = 44). A three-component partial least squares regression (PLSR) model using raw NIR spectra yielded a root mean square error of prediction (RMSEP) of 0.80% MC; a four component PLSR model using scatter corrected spectra yielded a RMSEP of 0.57% MC. A simplified PLS model using seven selected wavelengths (1155, 1212, 1340, 1409, 1724, 1908, and 2249 nm) yielded a similar accuracy (RMSEP: 0.77% MC) which opens the possibility of creating cheaper NIR instruments. In conclusion, NIR diffuse reflectance spectroscopy appears to be suitable for rapid and reliable MC prediction in intact green coffee; no separate model for Arabica and Robusta species is needed.