Hyperspectral Imaging and Chemometrics for Authentication of Extra Virgin Olive Oil: A Comparative Approach with FTIR, UV-VIS, Raman, and GC-MS

Detection of camellia oil adulteration based on near-infrared spectroscopy and smartphone combined with deep learning and multimodal fusion

Camellia oil (CO) is known for its nutritional value and health benefits, but its high price makes it susceptible to adulteration. This study developed a binary adulteration system for CO in response to the adulteration of rapeseed oil (RO) into CO that been observed in the market. A total of 243 oil samples adulterated with various concentrations of RO were prepared. The spectral information of the adulterated oil samples was obtained using near-infrared (NIR) spectroscopy. Additionally, visual data obtained from smartphone-captured images and videos were analysed. Deep-learning models trained on video data reached the highest accuracy of 96.30 %. To improve detection accuracy, a multimodal approach was adopted by combing spectral and visual data. Generally, this study presented a novel method for detecting the authenticity of CO in real time, providing technical support to address increasingly serious food safety concerns and laying the foundation for future rapid online detection using smartphones.

Authentication of Edible Oil by Real-Time One Class Classification Modeling

Adulteration detection or authentication is considered a type of one-class classification (OCC) in chemometrics. An effective OCC model requires representative samples. However, it is challenging to collect representative samples from all over the world. Moreover, it is also very hard to evaluate the representativeness of collected samples. In this study, we blazed a new trail to propose an authentication method to identify adulterated edible oils without building a prediction model beforehand. An authentication method developed by real-time one-class classification modeling, and model population analysis was designed to identify adulterated oils in the market without building a classification model beforehand. The underlying philosophy of the method is that the sum of the absolute centered residual (ACR) of the good model built by only authentic samples is higher than that of the bad model built by authentic and adulterated samples. In detail, a large number of OCC models were built by selecting partial samples out of inspected samples using Monte Carlo sampling. Then, adulterated samples involved in the test of these good models were identified. Taking the inspected samples of avocado oils as an example, as a result, 6 out of 40 avocado oils were identified as adulterated and then validated by chemical markers. The successful identification of avocado oils adulterated with soybean oil, corn oil, or rapeseed oil validated the effectiveness of our method. The proposed method provides a novel idea for oils as well as other high-value food adulteration detection.

Quantitative analysis of β-carotene and unsaturated fatty acids in blended olive oil via Raman spectroscopy combined with model prediction

Blended vegetable oil is considered to be a valuable product in the market owing to favourable taste and nutritional composition. The quantification of its contents has notable implications for protecting food safety and consumer interests. Thus, a rapid and non-destructive method is needed to analyse the composition of blended oil. This study established an analytical method combining Raman spectroscopy and prediction models to determine the content of olive oil in a mixture. Competitive adaptive reweighted sampling was employed to select feature bands attributed to β-carotene and unsaturated fatty acids. Various models were used to calculate the mixture proportion, and the importance of characteristic peak intensity affecting the prediction was evaluated via grey relational analysis. The random forest model exhibited superior performance in quantitative analysis, with RMSE and R2 of 0.0447 and 0.9799, respectively. Overall, this approach was proven to effectively identify blended olive oils, exemplifying its potential in food authentication.

Integrating near-infrared hyperspectral imaging with machine learning and feature selection: Detecting adulteration of extra-virgin olive oil with lower-grade olive oils and hazelnut oil

Detecting adulteration in extra virgin olive oil (EVOO) is particularly challenging with oils of similar chemical composition. This study applies near-infrared hyperspectral imaging (NIR-HSI) and machine learning (ML) to detect EVOO adulteration with hazelnut, refined olive, and olive pomace oils at various concentrations (1%, 5%, 10%, 20%, 40%, and 100% m/m). Savitzky-Golay filtering, first and second derivatives, multiplicative scatter correction (MSC), standard normal variate (SNV), and their combinations were used to preprocess the spectral data, with Principal Component Analysis (PCA) reducing dimensionality. Classification was performed using Partial Least Squares-Discriminant Analysis (PLS-DA) and ML algorithms, including k-Nearest Neighbors (k-NN), Naïve Bayes, Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Networks (ANN). PLS-DA, k-NN, RF, SVM, NB, and ANN models achieved accuracy rates of 97.0-99.0%, 96.2-100%, 96.5-100%, 98.6-99.5%, 93.9-99.7%, and 99.2-100%, respectively, in discriminating between pure EVOO, adulterants, and adulterated oils. PLS-DA, RF, SVM, and ANN significantly outperformed Naïve Bayes (p < 0.05) in binary classification, with Matthews correlation coefficient (MCC) values exceeding 0.90. All the binary classifiers except Naïve Bayes, when coupled with SNV/MSC, Savitzky-Golay smoothing and derivatives, consistently achieved perfect scores (1.0) for accuracy, sensitivity, specificity, F1 score, precision, and MCC in distinguishing pure EVOO from adulterated oils. No significant differences (p > 0.05) in model performance were found between those using full spectra and those based on key variable selection. However, PLS-DA and ANN significantly outperformed k-NN, RF, and SVM (p < 0.05), with MCC values ranging from 0.95 to 1.00, indicating superior classification performance. These findings demonstrate that combining NIR-HSI with machine learning, along with key variable selection, potentially offers an effective, non-destructive solution for detecting adulteration in EVOO and combating fraud in the olive oil industry.

Quantitative Prediction of Acid Value of Camellia Seed Oil Based on Hyperspectral Imaging Technology Fusing Spectral and Image Features

Acid value (AV) serves as an important indicator to assess the quality of oil, which can be used to judge the deterioration of edible oil. In order to realize the quantitative prediction of the AV of camellia seed oil, which was made from camellia oleifolia, hyperspectral data of 168 camellia seed oil samples were collected using a hyperspectral imaging system, which were related to their AV content measured via classical chemical titration. On the basis of hyperspectral full wavelengths, characteristic wavelengths, and fusing spectral and image features, the quantitative prediction AV models for camellia seed oil were established. The results demonstrating the 2Der-SPA-GLCM-PLSR model fusing spectral and image features stood out as the optimal choices for the AV prediction of camellia seed oil, with the correlation coefficient of calibration set (Rc2) and the correlation coefficient of prediction set (Rp2) at 0.9698 and 0.9581, respectively. Compared with those of 2Der-SPA-PLSR, the Rc2 and Rp2 were improved by 2.11% and 2.57%, respectively. Compared with those of 2Der-PLSR, the Rc2 and Rp2 were improved by 5.02% and 5.31%, respectively. Compared with the model based on original spectrum, the Rc2 and Rp2 were improved by 32.63% and 40.11%, respectively. After spectral preprocessing, characteristic wavelength selection, and fusing spectral and image features, the correlation coefficient of the optimal AV prediction model was continuously improved, while the root mean square error was continuously decreased. The research demonstrated that hyperspectral imaging technology could precisely and quantitatively predict the AV of camellia seed oil and also provide a new environmental method for detecting the AV of other edible oils, which is conducive to sustainable development.

A modified sequential wavenumber selection-discriminant analysis with Bayesian optimization strategy for detection and identification of chia seed oil adulteration using Raman spectroscopy

The detection of oil fraud can be accomplished through the use of Raman spectroscopy, which is a potent analytical technique for identifying the adulteration of edible oils with inferior or less expensive oils. However, appropriate data reduction and classification methods are required to achieve high accuracy and reliability in the analysis of Raman spectra. In this study, data reduction algorithms such as principal component analysis (PCA) and modified sequential wavenumber selection (MSWS) were applied, along with discriminant analysis (DA) as a classifier for detecting oil fraud. The parameters of DA, such as the discriminant type, the amount of regularization, and the linear coefficient threshold, were optimized using Bayesian optimization. The methods were tested on a dataset of chia oil mixed with 5-40 % sunflower oil, which is a common form of fraud in the market. The results showed that MSWS-DA achieved 100 % classification accuracy, while PCA-DA achieved 91.3 % accuracy. Therefore, it was demonstrated that Raman spectroscopy combined with MSWS-DA and Bayesian optimization can effectively detect oil fraud with high accuracy and robustness.

Trends in authentication of edible oils using vibrational spectroscopic techniques

The authentication of edible oils has become increasingly important for ensuring product quality, safety, and compliance with regulatory standards. Some prevalent authenticity issues found in edible oils include blending expensive oils with cheaper substitutes or lower-grade oils, incorrect labeling regarding the oil's source or type, and falsely stating the oil's origin. Vibrational spectroscopy techniques, such as infrared (IR) and Raman spectroscopy, have emerged as effective tools for rapidly and non-destructively analyzing edible oils. This review paper offers a comprehensive overview of recent advancements in using vibrational spectroscopy for authenticating edible oils. The fundamental principles underlying vibrational spectroscopy are introduced and chemometric approaches that enhance the accuracy and reliability of edible oil authentication are summarized. Recent research trends highlighted in the review include authenticating newly introduced oils, identifying oils based on their specific origins, adopting handheld/portable spectrometers and hyperspectral imaging, and integrating modern data handling techniques into the use of vibrational spectroscopic techniques for edible oil authentication. Overall, this review provides insights into the current state-of-the-art techniques and prospects for utilizing vibrational spectroscopy in the authentication of edible oils, thereby facilitating quality control and consumer protection in the food industry.

Machine learning identification of edible vegetable oils from fatty acid compositions and hyperspectral images

Hyperspectral imaging analysis combined with machine learning was applied to identify eight edible vegetable oils, and its classification performance was compared with the chemical method based on fatty acid compositions. Furthermore, the degree of adulteration in vegetable oils was quantitatively investigated using machine learning-enabled hyperspectral approaches. The hyperspectral absorbance spectra of palm oil with a high degree of saturation were distinctly different from those of the other liquid oils. The flaxseed and olive oils exhibited the dominant hyperspectral intensities at 1170/1671 and 1212/1415 nm, respectively. Linear discriminant analysis demonstrated that two linear discriminants could explain a significant portion of the total variability, accounting for 96.0% (fatty acid compositions) and 98.9% (hyperspectral images). When the hyperspectral results were used as datasets for three machine learning models (decision tree, random forest, and k-nearest neighbor), several instances to incorrectly classify grapeseed and sunflower oils were detected, while olive, palm, and flaxseed oils were successfully identified. The machine learning models showed a great classification performance that exceeded 98.9% from the hyperspectral images of the vegetable oils, which was comparable to the fatty acid composition-based chemical method in identifying edible vegetable oils. In addition, the random forest model was the most effective in ascertaining adulteration levels in binary oil blends (R2 > 0.992 and RMSE < 2.75).

Screening Method for the Visual Discrimination of Olive Oil from Other Vegetable Oils by a Multispecies DNA Sensor

Olive oil is a prominent agricultural product which, in addition to its nutritional value and unique organoleptic characteristics, offers a variety of health benefits protecting against cardiovascular disease, cancer, and neurodegenerative diseases. The assessment of olive oil authenticity is an extremely important and challenging process aimed at protecting consumers and producers. The most frequent adulteration involves blending with less expensive and readily available vegetable/seed oils. The methods for adulteration detection, whether based on changes in metabolite profiles or based on DNA markers, require advanced and expensive instrumentation combined with powerful chemometric and statistical tools. To this end, we present a simple, multiplex, and inexpensive screening method based on the development of a multispecies DNA sensor for sample interrogation with the naked eye. It is the first report of a DNA sensor for olive oil adulteration detection with other plant oils. The sensor meets the 2-fold challenge of adulteration detection, i.e., determining whether the olive oil sample is adulterated and identifying the added vegetable oil. We have identified unique, nucleotide variations, which enable the discrimination of seven plant species (olive, corn, sesame, soy, sunflower, almond, and hazelnut). Following a single PCR step, a 20 min multiplex plant-discrimination reaction is performed, and the products are applied directly to the sensing device. The plant species are visualized as red spots using functionalized gold nanoparticles as reporters. The spot position reveals the identity of the plant species. As low as <5-10% of adulterant was detected with particularly good reproducibility and specificity.

Rapid and Low-Cost Quantification of Adulteration Content in Camellia Oil Utilizing UV-Vis-NIR Spectroscopy Combined with Feature Selection Methods

This study aims to explore the potential use of low-cost ultraviolet-visible-near infrared (UV-Vis-NIR) spectroscopy to quantify adulteration content of soybean, rapeseed, corn and peanut oils in Camellia oil. To attain this aim, test oil samples were firstly prepared with different adulterant ratios ranging from 1% to 90% at varying intervals, and their spectra were collected by an in-house built experimental platform. Next, the spectra were preprocessed using Savitzky-Golay (SG)-Continuous Wavelet Transform (CWT) and the feature wavelengths were extracted using four different algorithms. Finally, Support Vector Regression (SVR) and Random Forest (RF) models were developed to rapidly predict adulteration content. The results indicated that SG-CWT with decomposition scale of 25 and the Iterative Variable Subset Optimization (IVSO) algorithm can effectively improve the accuracy of the models. Furthermore, the SVR model performed best for predicting adulteration of camellia oil with soybean oil, while the RF models were optimal for camellia oil adulterated with rapeseed, corn, or peanut oil. Additionally, we verified the models' robustness by examining the correlation between the absorbance and adulteration content at certain feature wavelengths screened by IVSO. This study demonstrates the feasibility of using low-cost UV-Vis-NIR spectroscopy for the authentication of Camellia oil.