Enhancing Coffee Leaf Rust Detection Using DenseNet201: A Comprehensive Analysis of the Mbozi and Public Datasets in Songwe, Tanzania

English

https://doi.org/10.51867/ajernet.6.1.17

Authors

Keywords:

Coffee Leaf Rust (CLR) Detection, Dataset Quality, DenseNet201, Image Quality, Machine Learning (ML)

Abstract

Coffee Leaf Rust (CLR) is a worldwide devastating fungal disease that threatens coffee production, upsetting economic and farmers' livelihoods. Traditional methods of detecting CLR heavily rely on using machine-learning (ML) models trained through weakly collected datasets and physical inspection which is tedious, time-consuming, and subject to human error. This study explores the performance of the DenseNet201 model using three datasets: Mbozi, Public, and Combined (a merger of Mbozi and Public datasets). Machine Learning Theory guided this research. The study objective is to assess the influence of dataset quality on CLR detection, analyze Mbozi and Public datasets using DenseNet201, and enhance robustness by merging the two datasets. A study on coffee leaf rot (CLR) severity was conducted using systematic sampling techniques. Leaves from multiple coffee farms were collected, representing different levels of infection. The Mbozi dataset, sourced from high-resolution images captured from Tanzania's Songwe coffee plantations, was analyzed for quality under controlled conditions, including environmental factors, image clarity, resolution, labeling consistency, and class balance, based on data completeness, image quality score, visual inspection, and model performance. DenseNet201 was trained and validated on each dataset achieving its highest accuracy with the Mbozi dataset at 98.72% and a validation accuracy of 97.65%, demonstrating the importance of consistent image quality and accurate annotations. In contrast, the public dataset suffered from inconsistencies in resolution and labeling, resulting in a lower training and validation accuracy of 96.86% and 96.42% respectively. The Combined dataset, which integrated the strengths of both datasets, exhibited a stronger generalization with an accuracy of 97.48% and validation accuracy of 97.49%, balancing the need for high-quality images with environmental variability. The study shows improved CLR detection speed and accuracy due to high-quality and consistently labeled images from the Mbozi dataset. It recommends future models integrate regionally relevant and high-resolution datasets for robust performance in real-world agricultural conditions, providing coffee farmers with timely disease intervention tools for better production management and economic stability in coffee-growing regions.

Dimensions

Al-Rashidi, J. F., Al-Enazi, L. A., Al-Mutairi, R. F., Al-Dukhayil, S. Y., Al-Abas, W. A., & Ibrahim, D. M. (2024). Coffee leaf diseases quadruple classifier (CLQC) model using deep learning. In A. Shaikh, A. Alghamdi, Q. Tan, & I. M. M. El Emary (Eds.), Advances in emerging information and communication technology. ICIEICT 2023. Signals and Communication Technology. Springer, Cham. https://doi.org/10.1007/978-3-031-53237-5_14 DOI: https://doi.org/10.1007/978-3-031-53237-5_14

Anania, P., & Nade, P. (2020). In the quest for semi-industrialized economy: Strategies for agricultural-based industrialization through co-operatives in Tanzania. Liège (Belgium): CIRIEC International, Université de Liège. https://doi.org/10.25518/ciriec.wp202006 DOI: https://doi.org/10.25518/ciriec.wp202006

Arathi, B., & Dulhare, U. N. (2023). Classification of cotton leaf diseases using transfer learning-DenseNet-121. In A. B. Reddy, S. Nagini, V. E. Balas, & K. S. Raju (Eds.), Proceedings of Third International Conference on Advances in Computer Engineering and Communication Systems. Lecture Notes in Networks and Systems, 612. Springer, Singapore. https://doi.org/10.1007/978-981-19-9228-5_33 DOI: https://doi.org/10.1007/978-981-19-9228-5_33

Bakr, M., Abdel-Gaber, S., Nasr, M., & Hazman, M. (2022). DenseNet based model for plant diseases diagnosis. European Journal of Electrical Engineering and Computer Science, 6(5), 1-9. https://doi.org/10.24018/ejece.2022.6.5.458 DOI: https://doi.org/10.24018/ejece.2022.6.5.458

Banothu, S., Madhavi, K., Madan Kumar, K. M. V., Gajula, R., Mallikarjuna Rao, C., Dixit, S., & Chhetri, A. (2024). Plant disease identification and pesticides recommendation using DenseNet. Cogent Engineering, 11(1), 45-49. https://doi.org/10.1080/23311916.2024.2353080 DOI: https://doi.org/10.1080/23311916.2024.2353080

Cabrera, A., Avramidis, K., & Narayanan, S. (2024). Early detection of coffee leaf rust through convolutional neural networks trained on low-resolution images. arXiv preprint, arXiv:2407.14737.

Chavarro, A. F., Renza, D., & Ballesteros, D. M. (2023). Influence of hyperparameters in deep learning models for coffee rust detection. Applied Sciences, 13(7), 4565. https://doi.org/10.3390/app13074565 DOI: https://doi.org/10.3390/app13074565

Gichuru, E., Alwora, G., Gimase, J., & Kathurima, C. (2021). Coffee leaf rust (Hemileia vastatrix) in Kenya-A review. Agronomy, 11(12), 2590. https://doi.org/10.3390/agronomy11122590 DOI: https://doi.org/10.3390/agronomy11122590

Hitimana, E., Sinayobye, O. J., Ufitinema, J. C., Mukamugema, J., Rwibasira, P., Murangira, T., Masabo, E., Chepkwony, L. C., Kamikazi, M. C. A., Uwera, J. A. U., Mvuyekure, S. M., Bajpai, G., & Ngabonziza, J. (2023). An intelligent system-based coffee plant leaf disease recognition using deep learning techniques on Rwandan Arabica dataset. Technologies, 11(5), 116. https://doi.org/10.3390/technologies11050116 DOI: https://doi.org/10.3390/technologies11050116

Jayaprakash, K., & Balamurugan, S. P. (2021). Analysis of plant disease detection and classification models: A computer vision perspective. Journal of Computational and Theoretical Nanoscience, 17(12), 5422-5428. https://doi.org/10.1166/jctn.2020.9435 DOI: https://doi.org/10.1166/jctn.2020.9435

Jepkoech, J., Mugo, D. M., Kenduiywo, B. K., & Too, E. C. (2021). Arabica coffee leaf images dataset for coffee leaf disease detection and classification. Data in Brief, 36. https://doi.org/10.1016/j.dib.2021.107142 DOI: https://doi.org/10.1016/j.dib.2021.107142

Ju, C., Chen, C., Li, R., Zhao, Y., Zhong, X., Sun, R., Liu, T., & Sun, C. (2023). Remote sensing monitoring of wheat leaf rust based on UAV multispectral imagery and the BPNN method. Food and Energy Security, 12, e477. https://doi.org/10.1002/fes3.477 DOI: https://doi.org/10.1002/fes3.477

Kiwelu, L., Damas, P., & Mpenda, Z. (2021). Factors influencing adoption of improved coffee varieties among smallholder farmers in Mbinga and Mbozi districts. International Journal of Agricultural Economics, 6(1), 21-32. https://doi.org/10.21203/rs.3.rs-181896/v1 DOI: https://doi.org/10.21203/rs.3.rs-181896/v1

Koutouleas, A. (2023). Coffee leaf rust: Wreaking havoc in coffee production areas across the tropics. Plant Health Cases, 2(23), 23-28.

https://doi.org/10.1079/planthealthcases.2023.0005 DOI: https://doi.org/10.1079/planthealthcases.2023.0005

Leandro, C. A., de Brito Silva, L., & Faulin, M. (2021). Artificial intelligence for detection and quantification of rust and leaf miner in coffee crop. arXiv e-prints, arXiv:2103.

Lyimo, D. A., Lakshmi Narasimhan, V., & Mbero, Z. A. (2021). Sensitivity analysis of coffee leaf rust disease using three deep learning algorithms. 2021 IEEE AFRICON. https://doi.org/10.1109/AFRICON51333.2021.9571007 DOI: https://doi.org/10.1109/AFRICON51333.2021.9571007

Magomba, L., & Ng'atigwa, F. (2024). Assessment of print media coverage of coffee crop farming in Tanzania: A case of Daily News and The Citizen newspapers. International Journal of Multidisciplinary Research and Growth Evaluation, 2(4), 275-287.

Mbwambo, S., Maro, G., Monyo, H., & Mosi, E. (2020). Towards expansion of Coffea canephora production in Tanzania: The land suitability perspective. World Journal of Agricultural Research, 8(2), 52-56. https://doi.org/10.12691/wjar-8-2-5

Mengistu, A. D., Alemayehu, D. M., & Mengistu, S. G. (2016). Ethiopian coffee plant diseases recognition based on imaging and machine learning techniques. International Journal of Database Theory and Application, 9(4), 79-88. https://doi.org/10.14257/ijdta.2016.9.4.07 DOI: https://doi.org/10.14257/ijdta.2016.9.4.07

Montalbo, F. J. P., & Hernandez, A. A. (2020, February). An optimized classification model for Coffea liberica disease using deep convolutional neural networks. In 2020 16th IEEE International Colloquium on Signal Processing & Its Applications (CSPA) (pp. 213-218).

https://doi.org/10.1109/CSPA48992.2020.9068683 DOI: https://doi.org/10.1109/CSPA48992.2020.9068683

Muharromah, M. D., Kristiana, A. I., Slamin, S., Dafik, D., Agustin, I. H., & Baihaki, R. I. (2024, July). The analysis of the implementation of convolutional neural network architectures for coffee leaf disease image classification. In AIP Conference Proceedings (Vol. 3176, No. 1). AIP Publishing. https://doi.org/10.1063/5.0225425 DOI: https://doi.org/10.1063/5.0225425

Nawaz, M., Nazir, T., Javed, A., Amin, S. T., Jeribi, F., & Tahir, A. (2024). CoffeeNet: A deep learning approach for coffee plant leaves disease recognition. Expert Systems with Applications, 237, 121481. doi: https://doi.org/10.1016/j.eswa.2023.121481. DOI: https://doi.org/10.1016/j.eswa.2023.121481

Novtahaning, D., Shah, H. A., & Kang, J.-M. (2022). Deep learning ensemble-based automated and high-performing recognition of coffee leaf disease. Agriculture, 12(11), 1909. https://doi.org/10.3390/agriculture12111909. DOI: https://doi.org/10.3390/agriculture12111909

Otieno, H. M., Alwenge, B. A., & Okumu, O. O. (2019). Coffee production challenges and opportunities in Tanzania: The case of coffee farmers in Iwindi, Msia, and Lwati villages in Mbeya Region. Asian Journal of Agricultural and Horticultural Research, 3(2), 1-14. https://doi.org/10.9734/ajahr/2019/v3i229993 DOI: https://doi.org/10.9734/ajahr/2019/v3i229993

Pham, T. C., Nguyen, V. D., Le, C. H., Packianather, M., & Hoang, V.-D. (2023). Artificial intelligence-based solutions for coffee leaf disease classification. IOP Conference Series: Earth and Environmental Science, 1278, 012004. https://doi.org/10.1088/1755-1315/1278/1/012004 DOI: https://doi.org/10.1088/1755-1315/1278/1/012004

Prince, R. H., Abdul, A. M., Hasibul, I. P., Shafiun, M., Md Nahiduzzaman, A. K., & Mohamed, A. A. (2024). CSXAI: A lightweight 2D CNN-SVM model for detection and classification of various crop diseases with explainable AI visualization. Frontiers in Plant Science, 15(4), 45-49. https://doi.org/10.3389/fpls.2024.1412988 DOI: https://doi.org/10.3389/fpls.2024.1412988

Pujiastuti, E. S., Page, N. R. H., Elisabeth, S. P., Rio, S., Susana, T., & Su, T. (2023). Increasing farmers' knowledge and action in agroforestry-based coffee cultivation. Journal of Research in Environmental and Earth Sciences, 9(2), 103-108.

Ruben, R., Catherine, A., Faycal, B., David, M., & Youri, D. (2018). Coffee value chain analysis in the southern highlands. Wageningen University and Research.

Soares, A. d. S., Vieira, B. S., Bezerra, T. A., Martins, G. D., & Siquieroli, A. C. S. (2022). Early detection of coffee leaf rust caused by Hemileia vastatrix using multispectral images. Agronomy, 2(22), 12, 2911. https://doi.org/10.3390/agronomy12122911 DOI: https://doi.org/10.3390/agronomy12122911

Tan, C., Sun, F., Kong, T., Zhang, W., Yang, C., & Liu, C. (2018). A survey on deep transfer learning. In: Kůrková, V., Manolopoulos, Y., Hammer, B., Iliadis, L., & Maglogiannis, I. (Eds.). Artificial Neural Networks and Machine Learning - ICANN 2018. Lecture Notes in Computer Science, Vol. 11141. Springer, Cham. https://doi.org/10.1007/978-3-030-01424-7_27. DOI: https://doi.org/10.1007/978-3-030-01424-7_27

Velásquez, D., Sánchez, A., Sarmiento, S., Velásquez, C., Toro, M., Montoya, E., Trefftz, H., Maiza, M., & Sierra, B. (2021). A cyber-physical data collection system integrating remote sensing and wireless sensor networks for coffee leaf rust diagnosis. Sensors, 21(16), 5474. https://doi.org/10.3390/s21165474 DOI: https://doi.org/10.3390/s21165474

Yebasse, M., Shimelis, B., Warku, H., Ko, J., & Cheoi, K. J. (2021). Coffee disease visualization and classification. Plants, 10(6), 1257. https://doi.org/10.3390/plants10061257 DOI: https://doi.org/10.3390/plants10061257

Published

2025-01-18

How to Cite

Karia, A. J., Ally, J. S., & Leonard, S. (2025). Enhancing Coffee Leaf Rust Detection Using DenseNet201: A Comprehensive Analysis of the Mbozi and Public Datasets in Songwe, Tanzania: English. African Journal of Empirical Research, 6(1), 171–188. https://doi.org/10.51867/ajernet.6.1.17