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...

Full description

Saved in:
Bibliographic Details
Main Authors: Derick Malavi, Katleen Raes, Sam Van Haute
Format: Article
Language:English
Published: Elsevier 2024-01-01
Series:Current Research in Food Science
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2665927124002399
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850245212832530432
author Derick Malavi
Katleen Raes
Sam Van Haute
author_facet Derick Malavi
Katleen Raes
Sam Van Haute
author_sort Derick Malavi
collection DOAJ
description 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.
format Article
id doaj-art-0f434df2663a4deeb6be3cf47baa2830
institution OA Journals
issn 2665-9271
language English
publishDate 2024-01-01
publisher Elsevier
record_format Article
series Current Research in Food Science
spelling doaj-art-0f434df2663a4deeb6be3cf47baa28302025-08-20T01:59:31ZengElsevierCurrent Research in Food Science2665-92712024-01-01910091310.1016/j.crfs.2024.100913Integrating near-infrared hyperspectral imaging with machine learning and feature selection: Detecting adulteration of extra-virgin olive oil with lower-grade olive oils and hazelnut oilDerick Malavi0Katleen Raes1Sam Van Haute2Department of Food Technology, Safety and Health, Faculty of Bioscience Engineering, Ghent University, Coupure Links 653, 9000, Ghent, Belgium; Center for Food Biotechnology and Microbiology, Ghent University Global Campus, 119, Songdomunhwa-Ro, Yeonsu-Gu, Incheon, 21985, South KoreaDepartment of Food Technology, Safety and Health, Faculty of Bioscience Engineering, Ghent University, Coupure Links 653, 9000, Ghent, BelgiumDepartment of Food Technology, Safety and Health, Faculty of Bioscience Engineering, Ghent University, Coupure Links 653, 9000, Ghent, Belgium; Center for Food Biotechnology and Microbiology, Ghent University Global Campus, 119, Songdomunhwa-Ro, Yeonsu-Gu, Incheon, 21985, South Korea; Corresponding author. Department of Food Technology, Safety and Health, Faculty of Bioscience Engineering, Ghent University, Coupure Links 653, 9000, Ghent, Belgium.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.http://www.sciencedirect.com/science/article/pii/S2665927124002399Machine learningVariable selectionExtra-virgin olive oil (EVOO)AdulterationAuthenticationClassification models
spellingShingle Derick Malavi
Katleen Raes
Sam Van Haute
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
Current Research in Food Science
Machine learning
Variable selection
Extra-virgin olive oil (EVOO)
Adulteration
Authentication
Classification models
title 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
title_full 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
title_fullStr 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
title_full_unstemmed 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
title_short 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
title_sort 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
topic Machine learning
Variable selection
Extra-virgin olive oil (EVOO)
Adulteration
Authentication
Classification models
url http://www.sciencedirect.com/science/article/pii/S2665927124002399
work_keys_str_mv AT derickmalavi integratingnearinfraredhyperspectralimagingwithmachinelearningandfeatureselectiondetectingadulterationofextravirginoliveoilwithlowergradeoliveoilsandhazelnutoil
AT katleenraes integratingnearinfraredhyperspectralimagingwithmachinelearningandfeatureselectiondetectingadulterationofextravirginoliveoilwithlowergradeoliveoilsandhazelnutoil
AT samvanhaute integratingnearinfraredhyperspectralimagingwithmachinelearningandfeatureselectiondetectingadulterationofextravirginoliveoilwithlowergradeoliveoilsandhazelnutoil