A Comparative Analysis of Deep Learning Architectures for The Classification of Madura Sliced Tobacco
Abstract
The manual grading of tobacco, a critical process determining its market value, is inherently subjective, labor-intensive, and prone to inconsistency. This study investigates the application of deep learning to automate the quality classification of Madura sliced tobacco, a high-value agricultural commodity. A novel, high-fidelity dataset was created, comprising 1,065 high-resolution images captured under standardized lighting and environmental conditions. An expert tobacco sorter with over five years of professional experience meticulously labeled these images into four distinct quality grades: Grade A (premium), Grade B (medium), Grade C (lower), and Grade X (waste). This research presents a rigorous comparative analysis of four pre-trained deep learning architectures MobileNetV3-Small, ResNet18, MobileViTV2, and EfficientNet-B0 fine-tuned for this specific classification task. Employing a 5-fold cross-validation methodology, the models were evaluated on their ability to accurately classify the tobacco grades. The experimental results reveal that the lightweight MobileNetV3-Small architecture achieved the highest mean test accuracy of 56.87%±2.71%. A detailed error analysis indicated that all models performed well on the majority classes but struggled significantly with the underrepresented minority classes, a challenge attributed to the dataset's severe class imbalance. This study validates the potential of lightweight deep learning models for automating tobacco classification, offering a promising pathway toward enhanced objectivity and efficiency in the agricultural industry. Furthermore, it establishes a crucial benchmark and identifies class imbalance as the primary obstacle to be addressed in future research for developing a field-deployable system.
Downloads
References
M. Zhang, Y. Liu, and Z. Chen, “Vision Transformer-Based Lightweight Models for Mobile Image Classification,” Pattern Recognit. Lett., vol. 181, pp. 14–22, 2025.
R. Yadav and S. Chauhan, “Precision Agriculture Using Machine Learning and IoT Technologies: A Review,” Agric. Syst., vol. 211, p. 103746, 2025.
X. Cao, Y. Liu, T. Zhang, and H. Wang, “Smart Farming Applications Based on Deep Learning: A Review of Current Trends and Future Directions,” Comput. Electron. Agric., vol. 225, p. 108932, 2025.
M. Sykes, D. Johnson, and R. Patel, “Deep Convolutional Neural Networks for Agricultural Product Grading: A Review,” Biosyst. Eng., vol. 238, pp. 45–62, 2024.
L. Hashemi-Beni and A. Gebrehiwot, “Deep Learning Approaches for Agricultural Image Analysis: A Review,” Sensors, vol. 20, no. 18, p. 5238, 2020, doi: 10.3390/s20185238.
R. Maximilliano and A. Rachmat, “Efficient Deep Learning Models for Classification of Indonesian Agricultural Products,” Journal of Advanced Computing Systems, vol. 11, no. 3, pp. 201–214, 2025.
A. Al Sahili and M. Awad, “Transfer Learning in Computer Vision: A Comprehensive Survey,” International Journal of Machine Learning and Cybernetics, vol. 13, no. 6, pp. 1725–1747, 2022, doi: 10.1007/s13042-021-01336-0.
M. Farhan, U. Tariq, and A. Rehman, “Transfer Learning Strategies for Small Agricultural Datasets: A Systematic Review,” Artificial Intelligence in Agriculture, vol. 7, pp. 45–60, 2023.
T. T. Mai, P. H. Nguyen, and D. L. Tran, “Fine-Tuning Pre-Trained Convolutional Neural Networks for Precision Agriculture Tasks,” Agricultural Informatics, vol. 16, no. 1, pp. 33–44, 2025.
H. Hindarto, “Application of ResNet Architecture for Agricultural Product Classification,” Indonesian Journal of Artificial Intelligence Research, vol. 2, no. 1, pp. 21–30, 2023.
B. Mouhcine, A. Elhassouny, and F. Khelifi, “Lightweight Convolutional Neural Networks for Mobile-Based Plant Classification,” Expert Syst. Appl., vol. 235, p. 121234, 2024.
F. Zhang and L. Zhang, “Machine Vision-Based Tobacco Grading System Using Image Processing Techniques,” Comput. Electron. Agric., vol. 78, no. 1, pp. 32–40, 2011.
S. Tonmoy, M. Rahman, and M. Islam, “Hybrid CNN-Transformer Architectures for Efficient Visual Recognition in Smart Farming,” IEEE Transactions on Artificial Intelligence, vol. 6, no. 1, pp. 89–102, 2025.
R. Teixeira, M. Silva, and P. Costa, “Automated Quality Assessment of Agricultural Products Using Deep Learning Techniques,” Comput. Electron. Agric., vol. 200, p. 107205, 2023.
K. Min, J. Park, and S. Lee, “Data-Driven Smart Agriculture Systems: Integration of IoT and Deep Learning,” Comput. Electron. Agric., vol. 226, p. 109015, 2025.
Z. Al Sahili and M. Awad, “The power of transfer learning in agricultural applications: AgriNet,” Front. Plant Sci., vol. Volume 13-2022, 2022, [Online]. Available: https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.992700
H. Li, Y. Xu, W. Chen, and S. Wang, “MobileViTv2: Improved Mobile Vision Transformers for Efficient Image Classification,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 45, no. 12, pp. 14876–14889, 2023.
J. Ma, Y. Sun, and H. Zhang, “Deep Residual Learning for Agricultural Image Recognition: Performance Evaluation Study,” Applied Sciences, vol. 12, no. 4, p. 2134, 2022.
Q. Hu, Y. Zhao, and X. Li, “Residual Neural Networks for Crop Quality Grading Using Image Processing,” IEEE Access, vol. 12, pp. 55678–55689, 2024.
Y. Ma, T. Chen, J. Zheng, W. Wang, and S. Gao, “Research on Identification Method and Improvement of Various Crop Diseases based on Resnet18,” International Core Journal of Engineering, vol. 8, no. 6, pp. 115–124, 2022, doi: 10.6919/ICJE.202206_8(6).0017.
D. Hindarto, “Comparison of Detection with Transfer Learning Architecture ResNet18, ResNet50, ResNet101 on Corn Leaf,” Jurnal Teknologi Informasi Universitas Lambung Mangkurat (JTIULM), vol. 8, no. 2, pp. 41–48, 2023, doi: 10.20527/jtiulm.v8i2.174.
W. Maximilliano and N. Rachmat, “Comparative Analysis of MobileNetV3-Large and Small for Corn Leaf Disease Classification,” Brilliance: Research of Artificial Intelligence, vol. 5, no. 1, pp. 325–332, Jul. 2025, doi: 10.47709/brilliance.v5i1.6259.
N. Afarini and D. Hindarto, “Optimization of CNN + MobileNetV3 for Insect Identification: Toward High Accuracy,” Jurnal Teknologi Informasi Universitas Lambung Mangkurat (JTIULM), vol. 9, no. 1, pp. 21–28, 2024, doi: 10.20527/jtiulm.v9i1.199.
K. Mouhcine, N. Zrira, I. Elafi, I. Benmiloud, and H. A. Khan, “STEFF: Spatio-temporal EfficientNet for dynamic texture classification in outdoor scenes,” Heliyon, vol. 10, no. 3, Feb. 2024, doi: 10.1016/j.heliyon.2024.e25360.
M. R. Tonmoy, Md. M. Hossain, N. Dey, and M. F. Mridha, “MobilePlantViT: A Mobile-friendly Hybrid ViT for Generalized Plant Disease Image Classification,” ArXiv, p., 2025, doi: 10.48550/arXiv.2503.16628.
G. Li, Y. Wang, Q. Zhao, P. Yuan, and B. Chang, “PMVT: a lightweight vision transformer for plant disease identification on mobile devices,” Front. Plant Sci., vol. Volume 14-2023, 2023, [Online]. Available: https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2023.1256773
Y. Guo, H. Shi, A. Kumar, and L. Wang, “Advances in Transfer Learning for Image Classification: Methods and Applications,” Pattern Recognit., vol. 147, p. 110012, 2024.
L. Darling-Hammond, L. Flook, C. Cook-Harvey, B. Barron, and D. Osher, “Implications for Educational Practice of the Science of Learning and Development,” Appl. Dev. Sci., vol. 25, no. 2, pp. 97–140, 2021, doi: 10.1080/10888691.2018.1537791.
L. Gao, J. Chen, and Z. Li, “EfficientNet-Based Lightweight Models for Crop Disease Recognition in Resource-Limited Environments,” Comput. Electron. Agric., vol. 203, p. 107467, 2023.
D. H. Schunk and M. K. DiBenedetto, “Self-Efficacy and Human Motivation in Educational Contexts,” Contemp. Educ. Psychol., vol. 66, p. 101944, 2021.
J. K. Burgoon, L. K. Guerrero, and K. Floyd, “Nonverbal Communication in Instructional Contexts: A Review of Contemporary Research,” J. Nonverbal Behav., vol. 45, no. 3, pp. 251–279, 2021.
Y. Gulzar, “Fruit Image Classification Model Based on MobileNetV2 with Deep Transfer Learning Technique,” Sustainability, vol. 15, no. 3, 2023, doi: 10.3390/su15031906.
Mr. M. D. V. V. S. Swaroop, V. A. Lakshmi, M. V. Vardhan, and B. N. S. G. Babu, “Plant Disease Classification Using Transfer Learning with ResNet Architecture,” International Journal of Scientific Advances in Technology (IJSAT), vol. 16, no. 3, 2025, doi: 10.71097/IJSAT.v16.i3.6958.
R. Gunasekaran and M. Thambusamy, “Deep Learning-Based Improved ResNet Model for Accurate Detection of Jackfruit Leaf Diseases,” Current Agriculture Research Journal, vol. 13, no. 2, p., 2025, doi: 10.12944/CARJ.13.2.20.
C. Yang and J. Kong, “Deploying an Optimized MobileNetV3 Model for Tomato Leaf Disease Identification on Android,” in 2024 4th International Symposium on Computer Technology and Information Science (ISCTIS), 2024, pp. 680–685. doi: 10.1109/ISCTIS63324.2024.10698865.
R. D. Rajagopal, M. R. Varma, and H. Rajagopal, “Efficient Weed Detection in Agricultural Landscapes using DeepLabV3+ and MobileNetV3,” in ICAROB2025: International Conference on Artificial Life and Robotics, 2025, pp. 60–66. doi: 10.5954/ICAROB.2025.OS2-1.
M. Zhang et al., “Dual-Attention-Enhanced MobileViT Network: A Lightweight Model for Rice Disease Identification in Field-Captured Images,” Agriculture, vol. 15, no. 6, 2025, doi: 10.3390/agriculture15060571.
Copyright (c) 2026 Muhammad Thahiruddin, Siti Khotijah, Adib El Farras, Afton Izzul Hasan, Ali Ya'lu S
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
This is an open-access article distributed under the terms of the Creative Commons Attribution-ShareAlike 4.0 International License which permits unrestricted use, distribution, and reproduction in any medium. Users are allowed to read, download, copy, distribute, search, or link to full-text articles in this journal without asking by giving appropriate credit, provide a link to the license, and indicate if changes were made. All of the remix, transform, or build upon the material must distribute the contributions under the same license as the original.
