Detecting adulterants in tea using mid-infrared spectroscopy: a comparative study of deep learning and machine learning

W. Liu; Y. Lin; Y. Cai; Honghao Cai; Hui Ni

Author(s)

W. Liu Department of Physics, School of Science, Jimei University,Xiamen, Fujian Province,China;School of Materials Science and Engineering, Guilin University of Technology,Guangxi Province,China

Y. Lin Department of Physics, School of Science, Jimei University,Xiamen, Fujian Province,China;School of Electronic Science and Engineering, Xiamen University,Xiamen, Fujian Province,China

Y. Cai Department of Physics, School of Science, Jimei University,Xiamen, Fujian Province,China

Honghao Cai Department of Physics, School of Science, Jimei University, Xiamen, Fujian Province, China

Hui Ni Xiamen Ocean Vocational College, Xiamen, Fujian Province, China

Abstract

The detection of adulterants in tea using infrared spectroscopy has gained prominence. However, there has yet to be a systematic comparison of the performance of traditional machine learning methods versus deep learning in the context of modelling infrared data for tea quality. This study compares machine learning and deep learning for modeling spectral data. Machine learning methods like Random Forest, K-Nearest Neighbors (KNN), Support Vector Classification, and Gaussian Naive Bayes used the Successive Projections Algorithm (SPA) for feature extraction, while deep learning models automatically extracted features. SPA-KNN showed superior performance with 0.950 accuracy, 0.953 macroprecision, 0.950 macro-recall, and 0.950 macro-F1 score on the test set (n=80), processing in 1.33 seconds. Deep learning models such as ResNet achieved a lower accuracy of 0.688 and required a longer processing time of 335.60 seconds. This may be partly due to the limited generalization ability caused by the small sample size. Additionally, the complex structure of ResNet, such as its depth, may contribute to the longer processing time. This study offers insights for selecting appropriate methods in tea quality assessment. Machine learning methods, especially with spectral preprocessing and SPA-based feature extraction, remain effective, while deep learning may not excel in limited data scenarios due to its higher computational demands.

Publication Details

Page(s) 889-899

DOI DOI not available

Published Journal: Journal of Animal and Plant sciences, Volume: 35, Issue: 4, Year: 2025

Request the full text from PASTIC

Keywords

LSTM, food quality, Adulteration, Classification Algorithm, 1D CNN, Feature selection, Resnet

References

Al-Mokhalelati K.,Karabet F.,Allaf A.,Naddaf M.,Assfour B.,Al Lafi A. .2023 .Silicone oils aided fabrication of paraffin wax coated superhydrophobic sand: A spectroscopic study. Heliyon, 9(10) : .

Amsaraj S.,Mutturi S. .2023 .Rapid detection of sunset yellow adulteration in tea powder with variable selection coupled to machine learning tools using spectral data. J. Food Sci. Tech, 60(5) : 1530-1540.

Amsaraj S.,Mutturi S. .2023 .Classification and quantification of multiple adulterants simultaneously in black tea using spectral data coupled with chemometric analysis. J. Food Compos. Anal, 125 : .

Anderson N.,Walsh K.,Subedi P.,Hayes C. .2020 .Achieving robustness across season, location and cultivar for a NIRS model for intact mango fruit dry matter content. Postharvest Biol. Tec, 1112 : .

Banerjee S.,Chatterjee S. .2015 .Efficient extraction strategies of tea (Camellia sinensis) biomolecules. J. Food Sci. Tech, 52 : 3158-3168.

Bai X.,Zhang L.,Kang C.,Quan B.,Zheng Y.,Zhang X.,Xia T.,Wang M. .2022 .Near-infrared spectroscopy and machine learning-based technique to predict qualityrelated parameters in instant tea. Sci. Rep, 12(1) : 3833.

Chen S.,Wang Y.,Zhu Q.,Ni H.,Cai H. .2023 .Fast recognition of the harvest period of Porphyra haitanensis based on mid-infrared spectroscopy and chemometrics. J. Food Meas. Charact, 17(5) : 5487-5496.

Cai Y.,Lin Y.,Cai H.,Ni H. .2025 .Deep learning in vibrational spectroscopy: Benefits, limitations, and recent progress. J. Chin. Chem. Soc, 72(6) : 611-626.

Esteki S.,Nouroozi Z.,Shahsavari Z. .2016 .A fast and direct spectrophotometric method for the simultaneous determination of methyl paraben and hydroquinone in cosmetic products using successive projections algorithm. Int. J. Cosmet. Sci, 38(1) : 25-34.

Gao Z.,Lin Q.,He Q.,Cai H.,Ni H. .2024 .Rapid detection of spoiled apple juice using electrical impedance spectroscopy and data augmentation-based machine learning. Chiang Mai J. Sci, 51(5) : .

Hou Z.,Jin Y.,Gu Z.,Zhang R.,Su Z.,Liu S. .2024 .1H NMR spectroscopy combined with machinelearning algorithm for origin recognition of Chinese famous green tea Longjing tea. Foods, 13(17) : 2702.

Hu Y.,Sheng W.,Adade S. Y. S. S.,Wang J.,Li H.,Chen Q. .2025 .Comparison of machine learning and deep learning models for detecting quality components of vine tea using smartphone-based portable near-infrared device. Food Control, 174 : 111244.

Ingebrigtson D. N.,Smith A. L. .1954 .Infrared analysis of solids by potassium bromide pellet technique. Anal. Chem, 26(11) : 1765-1768.

Kamrul M. H.,Rahman M.,Robin M. R. I.,Hossain M. S.,Hasan M. H.,P. Paul. M. H. .2020 .A deep learning based approach on categorization of tea leaf. Paper presented at the Proceedings of the International Conference on Computing Advancements, : .

Le B. T. .2019 .Application of deep learning and near infrared spectroscopy in cereal analysis. Vib. Spectrosc, 106 : .

Li F.,Huang Y.,Wang X.,Wang D.,Fan M. .2023 .Surface‐ enhanced Raman scattering integrating with machine learning for green tea storage time identification. Luminescence, 38(3) : 302-307.

Li X.,Zhang Y.,He Y. .2016 .Rapid detection of talcum powder in tea using FT-IR spectroscopy coupled with chemometrics. Sci. Rep, 6(1) : 30313.

Li X.,Zhang Y.,He Y. .2008 .Study on detection of talcum powder in green tea based on fourier transform infrared (FTIR) transmission spectroscopy. Spectrosc. Spect. Anal, 1000(4) : 10-2428.

Liao W.,Cai H.,Ni H. .2024 .Sesame seed metabolism during germination under auxin: An in vivo NMR study. J. Plant Growth Regul, 44 : 2764-2777.

Lin Y.,Cai H.,Lin S.,Ni H. .2025 .Mid-infraredspectroscopy-based method for identifying single and multiple vegetable protein adulterants in whey protein. J. Appl. Spectrosc, 91 : 1378-1386.

Lin Y.,Cai Y.,Chen H.,Cai Y.,Lin Z.,Cai H.,Ni H. .2025 .Adaptive feedback cross-loop for preserving and robust spectral information optimization without spectral processing in fewshot learning. Meas. Sci. Technol, 36 : 1361.

Liu M.,Wu Q.,Wang X.,Chen Q.,Zhang Y.,Huang S.,Fang J .2008 .Validity and redundancy of spectral data in the detection algorithm of sucrose-doped content in tea. Spectrosc. Spect. Anal, 1000(11) : 10-2428.

Liu W.,Lin Y.,Liu C.,Cai H.,Ni H. .2025 .Rapid detection of common additives in tea for quality assurance via mid-infrared spectroscopy and machine learning. Acta Sci. Polon, 001277(1) : 27-45.

Luo R.,Popp J.,Bocklitz T. .2022 .Deep learning for Raman spectroscopy: A review. Analytica, 3(3) : 287-301.

Magnus I.,Virte M.,Thienpont H.,Smeesters H. .2021 .Combining optical spectroscopy and machine learning to improve food classification. Food Control, 130 : 108342.

Melssen W.,Smits J.,Buydens L.,Kateman G. .1994 .Using artificial neural networks for solving chemical problems: Part II. Kohonen self-organising feature maps and Hopfield networks. Chemom. Intell. Lab. Syst, 23(93) : E0036-4.

Peng Y.,Zheng C.,Guo S.,Gao F.,Wang X.,Du Z.,Gao F.,Su F.,Zhang W.,Yu W. .2023 .Metabolomics integrated with machine learning to discriminate the geographic origin of Rougui Wuyi rock tea. npj Sci. Food, 7(1) : .

Qiu J.,Lin Y.,Wu J.,Xiao Y.,Cai H.,Ni H. .2024 .Rapid beef quality detection using spectra preprocessing methods in electrical impedance spectroscopy and machine learning. Int. J. Food Sci. Tech, 59(3) : 1624-1634.

Smits W.,Melssen L.,Buydens G.,Kateman G. .1994 .Using artificial neural networks for solving chemical problems: Part I. Multi-layer feed-forward networks. Chemom. Intell. Lab. Syst, 22(93) : E0035-3.

Sun J.,Zhou X.,Hu Y.,Wu X.,Zhang X.,Wang X. .2019 .Visualizing distribution of moisture content in tea leaves using optimization algorithms and NIR hyperspectral imaging. Comput. Electron. Agric, 03 : 153-159.

Vuong Q. V.,Stathopoulos C. E.,Golding J. B.,Nguyen and P. D. Roach M. H. .2011 .Optimum conditions for the water extraction of L‐ theanine from green tea. J. Sep. Sci, 34(18) : 2468-2474.

Wang X.,Chen H.,Ji R.,Qin H.,Xu Q.,Wang T.,He Y.,Huang Z. .2025 .Detection of carmine in black tea based on UV-Vis absorption spectroscopy and machine learning. Food Anal. Methods, 18(2) : 149-160.

Wu D.,Xu Y.,Xu F.,Shao M.,Huang M. .2024 .Machine learning algorithms for in-line monitoring during yeast fermentations based on Raman spectroscopy. Vib. Spectrosc, 132 : .

Xiao Y.,Cai H.,Ni H. .2024 .Identification of geographical origin and adulteration of Northeast China soybeans by mid-infrared spectroscopy and spectra augmentation. J. Consum. Prot Food S, 19(1) : 99-111.

Yan W.,Fan X.,Chen H.,Wang C.,Qin X.,Jiang X. .2022 .Component spectra extraction and quantitative analysis for preservative mixtures by combining terahertz spectroscopy and machine learning. Spectrochim. Acta, Part A, 271 : .

Yang C.,Yin X.,Miao X.,Meng Z.,Liu Z.,Hu Z. .2020 .Rapid discrimination of adulteration in Radix Astragali combining diffuse reflectance mid-infrared Fourier transform spectroscopy with chemometrics. Spectrochim. Acta, Part A, 248 : .

Yao L.,Chen Y.,Lin H.,Wang L.,Shi P.,Zhang Y.,Song J.,Wang Y.,Dai Q.,Liu C. .2024 .Simultaneous determination of 27 pyrrolizidine alkaloids in tea by ultra-high performance liquid chromatography-tandem mass spectrometry with integrated QuEChERS purification. J. Tea Sci, 05(5) : 831-842.

Ye C.,Zhang W.,Chen H.,Song Y.,Zhang P.,Gao P. .2022 .Detection of pesticide residue level in grape using hyperspectral imaging with machine learning. Foods, 11(11) : 1609.

Zhang H.,Dai J.,Zhang Z.,Zheng B.,Song J.,Chen G.,Lin L.,Chen W.,Sun Y.,Huang Y. .2023 .A study on origin traceability of white tea (White Peony) based on near-infrared spectroscopy and machine learning algorithms. Foods, 12(3) : 499.

Zhang X.,Yang J.,Lin T.,Ying Y. .2021 .Food and agro-product quality evaluation based on spectroscopy and deep learning: A review. Trends Food Sci. Tech, 04 : 431-441.