Automated Defect Classification of Coffee Beans Using Deep-Stacking Ensemble Learning
Article Sidebar
Main Article Content
Abstract
Quality control in coffee production is essential for maintaining product standards and preserving market value. A key practice in this process is to identify and remove defective beans which ensures high-quality standards and enhances consumer experience. However, traditional methods of classifying coffee bean defect often rely on manual inspection which is labour-intensive, time-consuming and subject to human errors. As such, adopting image classification for coffee bean defects could improve accuracy and boosts operational efficiency. This study explores the effectiveness of stacking-based deep learning ensemble method for coffee bean defect classification. The methodology involves a performance study as a baseline approach from fourteen traditional machine learning algorithms, including Support Vector Machines (SVM) and Random Forest (RF), along with ten different feature extraction techniques, such as FOS and GLDS. Besides, twenty well-known deep learning architectures including ResNet50, ConvNeXt and EfficientNet were compared to fourteen lightweight models such as TinyNet and MobileNet. Additionally, the performance of stacking-based deep learning models is also analysed to optimise coffee bean defect classification. The results indicate that ConvNeXt achieved the highest testing accuracy at 72.94% across all DL architectures. Additionally, the stacking approach significantly improves classification performance as it achieved an accuracy improvement from 72.94% to 87.64%. This study contributes to a comprehensive benchmarking to evaluate a diverse range of machine learning and deep learning algorithms. It also highlights the effectiveness of the stacking ensemble model to enhance accuracy in coffee bean defect classification.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
Alamri, E. S., Altarawneh, G. A., Bayomy, H. M., & Hassanat, A. B. (2023). Machine learning classification of roasted Arabic coffee: Integrating color, chemical compositions, and antioxidants. Sustainability, 15(11561). https://doi.org/10.3390/su151511561
Arwatchananukul, S., Xu, D., Charoenkwan, P., Moon, S. A., & Saengraya, R. (2024). Implementing a deep learning model for defect classification in Thai Arabica green coffee beans. Smart Agricultural Technology, 9, 100680. https://doi.org/10.1016/j.atech.2024.100680
Balakrishnan, B. N., Jayakumari, A. N. K. M., Stephen, S. A., Venkitesan, P., & Raghavendra, V. (2024). Coffee bean graded based on deep net models. International Journal of Electrical and Computer Engineering (IJECE), 14(3), 3084–3093. https://doi.org/10.11591/ijece.v14i3.pp3084-3093
Bhoomika, & Verma, G. (2024). Enhancing coffee plant disease identification with EfficientNetB0 and deep learning. Proceedings of the International Conference on I-SMAC (IoT in Social, Mobile, Analytics, and Cloud) (I-SMAC 2024). IEEE. https://doi.org/10.1109/I-SMAC61858.2024.10714676
Hassan, E. (2024). Enhancing coffee bean classification: A comparative analysis of pre-trained deep learning models. Neural Computing and Applications, 36(9023–9052). https://doi.org/10.1007/s00521-024-09623-z
Khan, A. H., Iskandar, D. N. A., Mewada, H., & Sherazi, M. A. (2022). Ensemble learning of deep learning and traditional machine learning approaches for skin lesion segmentation and classification. Concurrency and Computation: Practice and Experience. https://doi.org/10.1002/cpe.6907
Karypidis, E., Mouslech, S. G., Skoulariki, K., & Gazis, A. (2022). Comparison analysis of traditional machine learning and deep learning techniques for data and image classification. WSEAS Transactions on Mathematics, 21(19). https://doi.org/10.37394/23206.2022.21.19
Kumar, V., Aydav, P. S. S., & Minz, S. (2024). Crop seeds classification using traditional machine learning and deep learning techniques: A comprehensive survey. SN Computer Science, 5(1031). https://doi.org/10.1007/s42979-024-03379-y
Motta, I. V. C., Vuillerme, N., Pham, H.-H., & de Figueiredo, F. A. P. (2025). Machine learning techniques for coffee classification: A comprehensive review of scientific research. Artificial Intelligence Review, 58(1), 15. https://doi.org/10.1007/s10462-024-11004-w
Nawaz, M., Nazir, T., Javed, A., Amin, S. T., Jeribi, F., & Tahir, A. (2024). CoffeeNet: A deep learning approach for coffee plant leaves diseases recognition. Expert Systems With Applications, 237, 121481. https://doi.org/10.1016/j.eswa.2023.121481
Ossani, P. C., Rossoni, D. F., Cirillo, M. Â., & Borém, F. M. (2020). Unsupervised classification of specialty coffees in homogeneous sensory attributes through machine learning. Coffee Science, 15, e151780. https://doi.org/10.25186/cs.v15i.1780
Pragathi, S. P., & Jacob, L. (2022). Review on image-based coffee bean quality classification: Machine learning approach. 2022 4th International Conference on Advances in Computing, Communication Control and Networking (ICAC3N), 706–711. https://doi.org/10.1109/ICAC3N56670.2022.10074124
Rangarajan, A. K., Purushothaman, R., Prabhakar, M., & Szczepański, C. (2021). Crop identification and disease classification using traditional machine learning and deep learning approaches. Journal of Engineering Research, 11(1B), 228–252. https://doi.org/10.36909/jer.11941
Ramamurthy, K., Chattopadhyay, S., & Thekkath, R. D. (2023). A novel deep learning architecture for disease classification in Arabica coffee plants. Concurrency and Computation: Practice and Experience. https://doi.org/10.1002/cpe.7625
Selvanarayanan, R., Rajendran, S., & Alotaibi, Y. (2023). Early detection of Colletotrichum Kahawae disease in coffee cherry based on computer vision techniques. Computer Modeling in Engineering & Sciences. https://doi.org/10.32604/cmes.2023.044084
Singh, M. K., & Kumar, A. (2024). Coffee leaf disease classification by using a hybrid deep convolution neural network. SN Computer Science, 5(618). https://doi.org/10.1007/s42979-024-02960-9
Thai, H.-D., Ko, H.-J., & Huh, J.-H. (2024). Coffee bean defects automatic classification: Realtime application adopting deep learning. IEEE Access. https://doi.org/10.1109/ACCESS.2024.3452552
Thakur, D., Gera, T., Aggarwal, A., Verma, M., Kaur, M., Singh, D., & Amoon, M. (2024). SUNet: Coffee leaf disease detection using hybrid deep learning model. IEEE Access. https://doi.org/10.1109/ACCESS.2024.3476211
Tsai, J.-J., Chang, C.-C., Huang, D.-Y., Lin, T.-S., & Chen, Y.-C. (2023). Analysis and classification of coffee beans using single coffee bean mass spectrometry with machine learning strategy. Food Chemistry, 426, 136610. https://doi.org/10.1016/j.foodchem.2023.136610
Velásquez, S., Franco, A. P., Peña, N., Bohórquez, J. C., & Gutiérrez, N. (2021). Classification of the maturity stage of coffee cherries using comparative feature and machine learning. Coffee Science, 16, e161710. https://doi.org/10.25186/.v16i.1710
Wang, P., Fan, E., & Wang, P. (2021). Comparative analysis of image classification algorithms based on traditional machine learning and deep learning. Pattern Recognition Letters, 141, 61–67. https://doi.org/10.1016/j.patrec.2020.11.024
Zhang, X., Chan, F. T. S., & Mahadevan, S. (2022). Explainable machine learning in image classification models: An uncertainty quantification perspective. Knowledge-Based Systems, 243, 108418. https://doi.org/10.1016/j.knosys.2022.108418
Jantawong, S. (2024). Image detection of GT pesticide test kit results in organic agricultural products using the CiRA-CORE deep learning model. International Journal of Science and Innovative Technology, 7(2)