Abstract

Tea fermentation classification aims to categorize tea leaves according to their fermentation states using image analysis techniques, which often involve handling increasingly complex datasets. Conventional classification models frequently underperform due to their limited ability to effectively identify and exploit the most discriminative features from image data. To address this limitation and improve classification accuracy across diverse image datasets, this study proposes a robust framework for feature extraction and feature selection. The framework is specifically designed to enhance the classification of tea fermentation images while maintaining applicability to other image-processing tasks. It consists of four main stages: image acquisition, preprocessing, feature extraction, and feature selection. Feature extraction is performed across three categories—colour, texture, and wavelet transform features. Feature selection begins with correlation analysis (CA) to remove redundant features, followed by linear discriminant analysis (LDA) and principal component analysis (PCA) for dimensionality reduction. The proposed model is evaluated using six machine learning (ML) classifiers and two deep learning (DL) models on both a tea fermentation dataset and the Corel-1K image dataset. Initially, 94 features are extracted and subsequently reduced to 24 using the proposed feature-selection strategy. The results demonstrate superior performance, with the random forest (RF) classifier achieving an accuracy of 98.31% and gradient boosting (GB) achieving 99.71%. The DL models consistently outperform traditional ML techniques, with the convolutional neural network (CNN) showing the best classification performance. The model effectively distinguishes between under-fermented and fermented tea leaves, although over-fermented leaves exhibit slightly higher misclassification rates. Overall, the proposed approach outperforms existing methods in terms of precision, F-measure, and accuracy, albeit with slightly increased computational complexity due to multimodal feature processing and hybrid feature selection. The proposed system offers a novel, adaptable, and effective framework for feature extraction and selection, ensuring improved performance in image-classification tasks. It provides a solid foundation for future research in automated tea fermentation classification and other image-classification applications.

Keywords

Tea fermentation classification, Image processing, Feature extraction, Feature selection, Machine learning, Deep learning.

Cite this article

Sulaikha CM, Somasundaram A. A multimodal feature extraction and hybrid selection framework for image-based tea leaf fermentation classification. International Journal of Advanced Technology and Engineering Exploration. 2025;12(133):1807-1838. DOI : 10.19101/IJATEE.2025.121220208

References

[1] Surya G, Anbarasan P. Tea tales: an analytical exploration of branding strategies and consumer trends in the tea industry. Journal of Agriculture and Food Research. 2025:1-9.

[2] Banerjee S, Tyagi PK. Exploring the booming tea tourist industry and unconventional tourism through the ritual of drinking tea in India. Journal of Ethnic Foods. 2024; 11(1):1-19.

[3] Keelery S. Number of direct and indirect jobs in the travel and tourism sector across India from Financial Year 2014 to 2023, with Projection for 2024. Statista Research Department. 2024.

[4] Hannan A. Plantation economy and regional development. In the smallholder tea economy and regional development: perspectives from India 2024 (pp. 1-24). Cham: Springer International Publishing.

[5] Jagadeesh MS, Vinay HT, Pavithra V, Abhishek GJ. India’s tea export potential: stirring up global trade opportunities. Journal of Experimental Agriculture International. 2024; 46:309-19.

[6] Aaqil M, Peng C, Kamal A, Nawaz T, Zhang F, Gong J. Tea harvesting and processing techniques and its effect on phytochemical profile and final quality of black tea: a review. Foods. 2023; 12(24):1-28.

[7] Chen W, Zan J, Yan L, Yuan H, Wang P, Jiang Y, et al. Improving the sensory quality of black tea by blending varieties during processing. Foods. 2025; 14(6):1-17.

[8] Xu S, Wang H, Zhao Y. Optimizing fermentation conditions: impact on tea flavor and quality. Journal of Tea Science Research. 2024; 5(14):249-61.

[9] Yang L, Luo X, Wang Q, Liu M, Yan J, Wang C, et al. Exploring the effect of different tea varieties on the quality of sichuan congou black tea based on metabolomic analysis and sensory science. Frontiers in Nutrition. 2025; 12:1-16.

[10] Liu P, Feng L, Chen J, Wang S, Wang X, Han Y, et al. Unlocking the secrets of oingzhuan tea: a comprehensive overview of processing, flavor characteristics, and health benefits. Trends in Food Science & Technology. 2024; 147:104450.

[11] Hoque A, Padhiary M, Prasad G, Tiwari A. Optimization of fermentation time, temperature, and tea bed thickness in CFM to enhance the biological composition of CTC black tea. Journal of The Institution of Engineers (India): Series A. 2025; 106(2):305-18.

[12] Wei Y, Wen Y, Huang X, Ma P, Wang L, Pan Y, et al. The dawn of intelligent technologies in tea industry. Trends in Food Science & Technology. 2024; 144:104337.

[13] Trigka M, Dritsas E. A comprehensive survey of deep learning approaches in image processing. Sensors. 2025; 25(2):1-46.

[14] Han Z, Ahmad W, Rong Y, Chen X, Zhao S, Yu J, et al. A gas sensors detection system for real-time monitoring of changes in volatile organic compounds during oolong tea processing. Foods. 2024; 13(11):1-15.

[15] Guo J, Zhang K, Adade SY, Lin J, Lin H, Chen Q. Tea grading, blending, and matching based on computer vision and deep learning. Journal of the Science of Food and Agriculture. 2025; 105(6):3239-51.

[16] Bhuma CM, Kongara R. A novel technique for image retrieval based on concatenated features extracted from big dataset pre-trained CNNs. International Journal of Image, Graphics and Signal Processing. 2023; 14(2):1-12.

[17] Sayed MS, Gad-elrab AA, Fathy KA, Raslan KR. A deep learning content-based image retrieval approach using cloud computing. Indonesian Journal of Electrical Engineering and Computer Science. 2023; 29(3):1577-89.

[18] Maji S, Bose S. CBIR using features derived by deep learning. ACM/IMS Transactions on Data Science (TDS). 2021; 2(3):1-24.

[19] Sharif U, Mehmood Z, Mahmood T, Javid MA, Rehman A, Saba T. Scene analysis and search using local features and support vector machine for effective content-based image retrieval. Artificial Intelligence Review. 2019; 52(2):901-25.

[20] Yousuf M, Mehmood Z, Habib HA, Mahmood T, Saba T, Rehman A, et al. A novel technique based on visual words fusion analysis of sparse features for effective content‐based image retrieval. Mathematical Problems in Engineering. 2018; 2018(1):1-13.

[21] Ahmed KT, Ummesafi S, Iqbal A. Content based image retrieval using image features information fusion. Information Fusion. 2019; 51:76-99

[22] Sarwar A, Mehmood Z, Saba T, Qazi KA, Adnan A, Jamal H. A novel method for content-based image retrieval to improve the effectiveness of the bag-of-words model using a support vector machine. Journal of Information Science. 2019; 45(1):117-35.

[23] Mehmood Z, Mahmood T, Javid MA. Content-based image retrieval and semantic automatic image annotation based on the weighted average of triangular histograms using support vector machine. Applied Intelligence. 2018; 48(1):166-81.

[24] Ghaleb MS, Ebied HM, Shedeed HA, Tolba MF. Image retrieval based on deep learning. Journal of System and Management Sciences. 2022; 12(2):477-96.

[25] Wang L, Xie J, Wang Q, Hu J, Jiang Y, Wang J, et al. Evaluation of the quality grade of congou black tea by the fusion of GC-E-nose, E-tongue, and E-eye. Food Chemistry: X. 2024; 23:1-9.

[26] Bhargava A, Bansal A, Goyal V, Shukla A. Machine learning & computer vision-based optimum black tea fermentation detection. Multimedia Tools and Applications. 2023; 82(28):43335-47.

[27] Li T, Lu C, Huang J, Chen Y, Zhang J, Wei Y, et al. Qualitative and quantitative analysis of the pile fermentation degree of pu-erh tea. LWT. 2023; 173:1-9.

[28] Zheng P, Solomon ASY, Rong Y, Zhao S, Han Z, Gong Y, et al. Online system for monitoring the degree of fermentation of oolong tea using integrated visible–near-infrared spectroscopy and image-processing technologies. Foods. 2024; 13(11):1-14.

[29] Kimutai G, Ngenzi A, Said RN, Kiprop A, Förster A. An optimum tea fermentation detection model based on deep convolutional neural networks. Data. 2020; 5(2):1-26.

[30] Yu D, Gu Y. A machine learning method for the fine-grained classification of green tea with geographical indication using a MOS-based electronic nose. Foods. 2021; 10(4):1-17.

[31] Cui Q, Yang B, Liu B, Li Y, Ning J. Tea category identification using wavelet signal reconstruction of hyperspectral imagery and machine learning. Agriculture. 2022; 12(8):1-16.

[32] An T, Yang C, Zhang J, Wang Z, Fan Y, Fan S, et al. Evaluation of the black tea taste quality during fermentation process using image and spectral fusion features. Fermentation. 2023; 9(10):1-15.

[33] Zhou Q, Dai Z, Song F, Li Z, Song C, Ling C. Monitoring black tea fermentation quality by intelligent sensors: comparison of image, e-nose and data fusion. Food Bioscience. 2023; 52:102454.

[34] Zhang B, Li Z, Song F, Zhou Q, Jin G, Raghavan V, et al. Discrimination of black tea fermentation degree based on multi-data fusion of near-infrared spectroscopy and machine vision. Journal of Food Measurement and Characterization. 2023; 17(4):4149-60.

[35] Ding Z, Yang C, Hu B, Guo M, Li J, Wang M, et al. Lightweight CNN combined with knowledge distillation for the accurate determination of black tea fermentation degree. Food Research International. 2024; 194:114929.

[36] Huda HS, Majid NB, Chen Y, Adnan M, Ashraf SA, Roszko M, et al. Exploring the ancient roots and modern global brews of tea and herbal beverages: a comprehensive review of origins, types, health benefits, market dynamics, and future trends. Food Science & Nutrition. 2024; 12(10):6938-55.

[37] Ren G, Wu R, Yin L, Zhang Z, Ning J. Description of tea quality using deep learning and multi-sensor feature fusion. Journal of Food Composition and Analysis. 2024; 126:105924.

[38] Hu Y, Chen W, Gouda M, Yao H, Zuo X, Yu H, et al. Fungal fermentation of Fuzhuan brick tea: a comprehensive evaluation of sensory properties using chemometrics, visible near-infrared spectroscopy, and electronic nose. Food Research International. 2024; 186:114401.

[39] Huang M, Tang Y, Tan Z, Ren J, He Y, Huang H. Detection of black tea fermentation quality based on optimized deep neural network and hyperspectral imaging. Infrared Physics & Technology. 2024; 143:105625.

[40] Huang Y, Zhao J, Zheng C, Li C, Wang T, Xiao L, et al. The fermentation degree prediction model for tieguanyin oolong tea based on visual and sensing technologies. Foods. 2025; 14(6):1-18.

[41] Zhu F, Yao H, Shen Y, Zhang Y, Li X, Shi J, et al. Information fusion of hyperspectral imaging and self-developed electronic nose for evaluating the degree of black tea fermentation. Journal of Food Composition and Analysis. 2025; 137:106859.

[42] Qi F, Bai S, Zou D, Tang Z. Discrimination of black tea fermentation degree with integrated attention mechanisms. Journal of Food Measurement and Characterization. 2025:1-20.

[43] Ouyang Q, Chang H, Fan Z, Ma S, Chen Q, Liu Z. Monitoring changes in constituents during black tea fermentation using snapshot multispectral imaging and 1D-CNN enhanced with data augmentation. Computers and Electronics in Agriculture. 2025; 237:1-9.

[44] Eissa AH, Khalik AA. Understanding color image processing by machine vision for biological materials. In structure and function of food engineering 2012. IntechOpen.

[45] Li H, Mukundan R, Boyd S. Novel texture feature descriptors based on multi-fractal analysis and LBP for classifying breast density in mammograms. Journal of Imaging. 2021; 7(10):1-21.

[46] Ahmed HM. Texture feature extraction using tamura descriptors and scale-invariant feature transform. Journal of Education & Science. 2023; 32(4):91-103.

[47] Gopalsamy A, Radha B, Haridas K. Prediction of neurodegenerative disease using brain image analysis with multilinear principal component analysis and quadratic discriminant analysis. International Journal of Advanced Technology and Engineering Exploration. 2022; 9(90):604-22.

[48] Rathnayaka RA, Jananey B, Maduranga HP, Sivananthawerl T, Amarasena PG, Frossard E, et al. Allometric models for estimating above-ground, below-ground and total biomass of tea (Camellia sinensis (L:) O. Tropical Agricultural Research. 2024; 35(4):275-87.

[49] Yu L, Liu H. Feature selection for high-dimensional data: a fast correlation-based filter solution. In proceedings of the 20th international conference on machine learning (ICML-03) 2003 (pp. 856-63).

[50] Afjal MI, Mondal MN, Mamun MA. Segmentation-based linear discriminant analysis with information theoretic feature selection for hyperspectral image classification. International Journal of Remote Sensing. 2023; 44(11):3412-55.

[51] Omuya EO, Okeyo GO, Kimwele MW. Feature selection for classification using principal component analysis and information gain. Expert Systems with Applications. 2021; 174:1-12.

[52] Kimutai G, Ngenzi A, Ngoga SR, Ramkat RC, Förster A. A data descriptor for black tea fermentation dataset. Data. 2021; 6(3):1-8.

[53] Bama SS, Ahmed MI, Saravanan A. A survey on performance evaluation measures for information retrieval system. International Research Journal of Engineering and Technology. 2015; 2(2):1015-20.

[54] Pfaehler E, Mesotten L, Zhovannik I, Pieplenbosch S, Thomeer M, Vanhove K, et al. Plausibility and redundancy analysis to select FDG‐PET textural features in non‐small cell lung cancer. Medical Physics. 2021; 48(3):1226-38.