Fusion of Electronic Nose and Computer Vision for Detecting Lampante Oil Adulteration in Extra Virgin Olive Oil Using Machine Learning Algorithms
Subject Areas : Multimedia Processing, Communications Systems, Intelligent SystemsMahsa Mirhoseini Moghaddam 1 , Mohammad reza Yamaghani 2 * , Adel Bakhshipour 3
1 - Ph.D. Student, Department of Computer Engineering, Lahijan Branch, Islamic Azad University, Lahijan, Iran
2 - Assistant Professor, Department of Computer Engineering, Lahijan Branch, Islamic Azad University, Lahijan, Iran
3 - Assistant Professor, Department of Biosystems Engineering, Faculty of Agricultural Sciences, University of Guilan, Rasht, Iran
Keywords: Adulteration, chemometrics, electronic nose, image processing, data fusion,
Abstract :
Introduction: Olive (Olea europaea L.), a key economic fruit species primarily cultivated in the Mediterranean region, is widely consumed as table olives and olive oil. Olive oil adulteration, a significant food fraud issue, typically involves mixing lower-quality oils with extra virgin olive oil (EVOO), compromising its quality and nutritional value. This study explores the use of electronic nose (e-nose) and computer vision (CV) systems to detect adulteration in EVOO.
Method: Four levels of adulteration (5%, 10%, 15%, and 20%) were simulated by blending lampante oil with pure EVOO. The aroma and appearance of the pure and adulterated samples were captured using an e-nose system equipped with 13 gas sensors and a CV apparatus. The collected signals and images were processed in MATLAB (R2021a, The MathWorks, USA) for feature extraction, resulting in 78 odor-related features (13 sensors × 6 features) and 9 color-related features. These were then used to train machine learning (ML) algorithms.
Results: Using e-nose data, the quadratic discriminant analysis (QDA) algorithm, combined with principal components analysis (PCA), we achieved 100% accuracy in distinguishing pure from adulterated EVOO. With CV data, the PCA-QDA model provided 98.50% accuracy. Furthermore, the partial least squares (PLS) algorithm predicted the percentage of lampante oil adulteration with determination coefficients (R²) of 0.8565 in the training phase and 0.7858 in the evaluation phase. These values dropped to 0.6983 and 0.4936, respectively, when only color features were used. However, by fusing aroma and color data, the prediction accuracy significantly improved, with R² values of 0.9266 for training and 0.9184 for evaluation.
Discussion: The findings demonstrate that combining e-nose and CV data with appropriate ML algorithms allows highly accurate detection of EVOO adulteration. Robustness tests, performed by intentionally altering the e-nose and CV data, confirmed the models' reliability. Given the e-nose and CV systems' capabilities and the achieved performance, this approach offers a fast, accurate, cost-effective, and non-destructive method for detecting adulteration in EVOO.
1. Sakar, E., et al., Genetic relationships among olive (Olea europaea L.) cultivars native to Turkey. Biochemical genetics, 2016. 54: p. 348-359.
2. FAOSTAT, Crop statistics. Food and Agriculture Organization of the United Nations (FAO). 2023.
3. Bendini, A., et al., Phenolic molecules in virgin olive oils: a survey of their sensory properties, health effects, antioxidant activity and analytical methods. An overview of the last decade. Molecules, 2007. 12(8): p. 1679-1719.
4. Bendini, A., et al., Sensory analysis of virgin olive oil. Olive Oil—Constituents, Quality, Health Properties and Bioconversions; InTech: Rijeka, Croatia, 2012: p. 109-130.
5. Roy, M. and B. Yadav, Electronic nose for detection of food adulteration: A review. Journal of Food Science and Technology, 2022: p. 1-13.
6. Mohammad‐Razdari, A., et al., Potential application of electronic nose coupled with chemometric tools for authentication assessment in tomato paste. Journal of Food Process Engineering, 2019. 42(5): p. e13119.
7. Lerma-García, M., et al., Metal oxide semiconductor sensors for monitoring of oxidative status evolution and sensory analysis of virgin olive oils with different phenolic content. Food Chemistry, 2009. 117(4): p. 608-614.
8. Mildner‐Szkudlarz, S. and H.H. Jeleń, Detection of olive oil adulteration with rapeseed and sunflower oils using mos electronic nose and SMPE‐MS. Journal of food quality, 2010. 33(1): p. 21-41.
9. Oates, M.J., et al., DFT based classification of olive oil type using a sinusoidally heated, low cost electronic nose. Computers and Electronics in Agriculture, 2018. 155: p. 348-358.
10. Tian, H., et al., Rapid identification and quantification of vegetable oil adulteration in raw milk using a flash gas chromatography electronic nose combined with machine learning. Food Control, 2023. 150: p. 109758.
11. Ayari, F., et al., Using an E‐nose machine for detection the adulteration of margarine in cow ghee. Journal of Food Process Engineering, 2018. 41(6): p. e12806.
12. Milanez, K.d.M. and M. Pontes, Classification of extra virgin olive oil and verification of adulteration using digital images and discriminant analysis. Analytical Methods, 2015. 7(20): p. 8839-8846.
13. Kadiroğlu, P. and F. Korel, Chemometric studies on zNose™ and machine vision technologies for discrimination of commercial extra virgin olive oils. Journal of the American Oil Chemists' Society, 2015. 92: p. 1235-1242.
14. Ram, T., et al., Olive oil content prediction models based on image processing. Biosystems engineering, 2010. 105(2): p. 221-232.
15. Moyano, M., et al., A comprehensive study on the colour of virgin olive oils and its relationship with their chlorophylls and carotenoids indexes (I): CIEXYZ non-uniform colour space. Food Research International, 2008. 41(5): p. 505-512.
16. Marchal, P.C., et al., Expert system based on computer vision to estimate the content of impurities in olive oil samples. Journal of Food Engineering, 2013. 119(2): p. 220-228.
17. Li, Q., et al., Machine learning technique combined with data fusion strategies: A tea grade discrimination platform. Industrial Crops and Products, 2023. 203: p. 117127.
18. Guo, M., et al., Spectral data fusion in nondestructive detection of food products: Strategies, recent applications, and future perspectives. Comprehensive Reviews in Food Science and Food Safety, 2024. 23(1): p. 1-23.
19. Kiani, S., S. Minaei, and M. Ghasemi-Varnamkhasti, Application of electronic nose systems for assessing quality of medicinal and aromatic plant products: A review. Journal of Applied Research on Medicinal and Aromatic Plants, 2016. 3(1): p. 1-9.
20. Khulal, U., et al., Nondestructive quantifying total volatile basic nitrogen (TVB-N) content in chicken using hyperspectral imaging (HSI) technique combined with different data dimension reduction algorithms. Food chemistry, 2016. 197: p. 1191-1199.
21. Wang, W., et al., Advances and challenges in medicinal plant breeding. Plant Science, 2020. 298: p. 110573.
22. Bakhshipour, A., A data fusion approach for nondestructive tracking of the ripening process and quality attributes of green Hayward kiwifruit using artificial olfaction and proximal hyperspectral imaging techniques. Food Science & Nutrition, 2023. 11(10): p. 6116-6132.
23. Asadi, M., et al., Predicting the quality attributes related to geographical growing regions in red-fleshed kiwifruit by data fusion of electronic nose and computer vision systems. BMC Plant Biology, 2024. 24(1): p. 13.
24. Pearce, T.C., et al., Handbook of machine olfaction: electronic nose technology. 2006: John Wiley & Sons.
25. Aghilinategh, N., M.J. Dalvand, and A. Anvar, Detection of ripeness grades of berries using an electronic nose. Food Science & Nutrition, 2020. 8(9): p. 4919-4928.
26. Yin, Y., et al., A selection method for feature vectors of electronic nose signal based on wilks Λ–statistic. Journal of Food Measurement and Characterization, 2014. 8: p. 29-35.
27. Nimsuk, N., Improvement of accuracy in beer classification using transient features for electronic nose technology. Journal of Food Measurement and Characterization, 2019. 13(1): p. 656-662.
28. Sanaeifar, A., et al., A data fusion approach on confocal Raman microspectroscopy and electronic nose for quantitative evaluation of pesticide residue in tea. Biosystems Engineering, 2021. 210: p. 206-222.
29. Haghbin, N., et al., Monitoring Botrytis cinerea infection in kiwifruit using electronic nose and machine learning techniques. Food and Bioprocess Technology, 2023. 16(4): p. 749-767.
30. Amirvaresi, A., et al., Comparison of near-infrared (NIR) and mid-infrared (MIR) spectroscopy based on chemometrics for saffron authentication and adulteration detection. Food Chemistry, 2021. 344: p. 128647.
31. Keshavarzi, Z., et al., Comparison of transmission FTIR and ATR spectra for discrimination between beef and chicken meat and quantification of chicken in beef meat mixture using ATR-FTIR combined with chemometrics. Journal of food science and technology, 2020. 57: p. 1430-1438.
32. Mishra, P., et al., New data preprocessing trends based on ensemble of multiple preprocessing techniques. TrAC Trends in Analytical Chemistry, 2020. 132: p. 116045.
33. Sarkar, S., et al., A comparative study of PLSR and SVM-R with various preprocessing techniques for the quantitative determination of soluble solids content of hardy kiwi fruit by a portable vis/NIR spectrometer. Foods, 2020. 9(8): p. 1078.
34. Muthahharah, I. and A. Juhari, A Cluster Analysis with Complete Linkage and Ward's Method for Health Service Data in Makassar City. Jurnal Varian, 2021. 4(2): p. 109-116.
35. Joshi, R., et al., Evaluation of benzene residue in edible oils using Fourier transform infrared (FTIR) spectroscopy. Korean Journal of Agricultural Science, 2019. 46(2): p. 257-271.
36. Maione, C., F. Barbosa Jr, and R.M. Barbosa, Predicting the botanical and geographical origin of honey with multivariate data analysis and machine learning techniques: A review. Computers and Electronics in Agriculture, 2019. 157: p. 436-446.
37. Sabouri, A., et al., Application of image processing and soft computing strategies for non-destructive estimation of plum leaf area. PLoS One, 2022. 17(7): p. e0271201.
38. Wang, F., et al., Applying different resampling strategies in machine learning models to predict head-cut gully erosion susceptibility. Alexandria Engineering Journal, 2021. 60(6): p. 5813-5829.
39. Jiang, H., Y. He, and Q. Chen, Qualitative identification of the edible oil storage period using a homemade portable electronic nose combined with multivariate analysis. Journal of the Science of Food and Agriculture, 2021. 101(8): p. 3448-3456.
40. Karami, H., M. Rasekh, and E. Mirzaee-Ghaleh, Qualitative analysis of edible oil oxidation using an olfactory machine. Journal of Food Measurement and Characterization, 2020. 14: p. 2600-2610.
41. Buratti, S., et al., E-nose, e-tongue and e-eye for edible olive oil characterization and shelf life assessment: A powerful data fusion approach. Talanta, 2018. 182: p. 131-141.
42. Karagiorgos, N., et al. An approach for estimating adulteration of virgin olive oil with soybean oil using image analysis, 2017 6th Int. in Conf. Mod. Circuits Syst. Technol. MOCAST. 2017.
43. Teixeira, G.G., et al., Application of a lab-made electronic nose for extra virgin olive oils commercial classification according to the perceived fruitiness intensity. Talanta, 2021. 226: p. 122122.
44. Wei, X., et al., Rapid detection of adulterated peony seed oil by electronic nose. Journal of food science and technology, 2018. 55: p. 2152-2159.
45. Men, H., et al., Data fusion of electronic nose and electronic tongue for detection of mixed edible-oil. Journal of Sensors, 2014. 2014: p. 1-7.
46. Bougrini, M., et al., Detection of adulteration in argan oil by using an electronic nose and a voltammetric electronic tongue. Journal of Sensors, 2014. 2014.
47. Marina, A., Y. Che Man, and I. Amin, Use of the SAW sensor electronic nose for detecting the adulteration of virgin coconut oil with RBD palm kernel olein. Journal of the American Oil Chemists' Society, 2010. 87(3): p. 263-270.