Crop Disease Prediction Using Deep Learning Algorithm

Abstract

Plant diseases pose a significant threat to global agricultural productivity, particularly affecting key crops like tomato and potato. Traditional disease detection methods are often slow, subjective, and labour-intensive, leading to delayed responses and increased crop losses. This study proposes a hybrid machine learning framework that integrates ResNet9 for classification and U-Net for segmentation to detect and localize leaf diseases in tomato and potato plants. A comprehensive dataset of over 22,500 images spanning 13 classes, including healthy and diseased samples, was compiled from multiple sources and pre-processed using image normalization, histogram equalization, and data augmentation techniques. The model was trained using a 70:20:10 data split and optimized through early stopping and cyclic learning rates. Evaluation metrics including accuracy, precision, recall, F1-score, and ROC-AUC were used to assess performance, with the proposed model achieving a remarkable accuracy of 96.3%, F1-score of 95.2%, and ROC-AUC of 97.1%. The use of U-Net enabled accurate segmentation of infected regions, improving model interpretability and trustworthiness. Confusion matrix analysis revealed minimal misclassifications, and visual tools such as saliency maps confirmed the model’s attention to disease-prone areas. Real-world testing demonstrated the system’s robustness across different environments and lighting conditions. Comparative results showed superior performance of the hybrid model over VGG16, EfficientNet-B0, and baseline CNNs in both accuracy and inference speed. This approach offers a scalable, real-time solution for automated plant disease detection and diagnosis, particularly suited for use in resource-constrained agricultural settings. The hybrid model not only supports early intervention and precision agriculture practices but also bridges the gap between advanced machine learning and practical farming needs.

Introduction

Agriculture is vital globally for economy and food security, but plant diseases threaten crop yield and quality, especially in tomato and potato plants. Traditional disease detection methods, relying on manual inspection, are subjective, time-consuming, and prone to errors. Laboratory tests are accurate but costly and impractical for large-scale field use.

Recent advances in Artificial Intelligence, particularly deep learning, offer promising automated solutions. This research develops a hybrid deep learning system combining ResNet9 for disease classification and U-Net for spatial segmentation of infected areas. This dual approach enhances accuracy and interpretability by not only identifying the disease but also localizing affected leaf regions.

A comprehensive dataset of 22,546 images across 13 classes of tomato and potato leaf diseases was assembled, pre-processed, and augmented to improve model robustness. The hybrid model achieved high accuracy (96.3%) and strong performance metrics, outperforming other models like VGG16 and EfficientNet-B0.

The study demonstrates that integrating classification and segmentation with proper data handling produces an efficient, interpretable, and scalable system for early plant disease detection. This approach supports better disease management, reduces crop losses, and promotes food security, addressing limitations of manual and traditional diagnostics.

Conclusion

This study has successfully developed and evaluated a hybrid machine learning model for the detection and localization of leaf diseases in tomato and potato plants. By integrating ResNet9 and U-Net architectures, the model combines high-accuracy classification capabilities with effective visual interpretability through segmentation. This dual approach not only enhances the technical performance of disease detection but also builds transparency and trust among end-users such as farmers and agronomists. The collection of a comprehensive and balanced dataset from multiple sources allowed for robust model training, which was further strengthened through pre-processing techniques such as data augmentation, histogram equalization, and noise filtering. These enhancements significantly contributed to the improved learning ability of the model by exposing it to diverse real-world scenarios. The classification component, powered by ResNet9, achieved high accuracy (96.3%), precision (94.8%), recall (95.6%), and F1-score (95.2%), indicating excellent performance across multiple disease categories. The segmentation component, managed by U-Net, accurately identified infected regions, enabling users to visually confirm the presence of disease. These results were supported by performance metrics and visual analysis tools such as saliency maps and confusion matrices, which highlighted the model\'s focus on actual disease symptoms rather than irrelevant image features. The inclusion of dimensionality reduction techniques and proper data splitting ensured minimal overfitting and better generalization, as evident from consistent validation scores and real-world test performance. Comparative analysis further confirmed that the hybrid model outperforms traditional architectures like VGG16 and EfficientNet-B0 in both accuracy and computational efficiency. Its light-weight architecture, fast inference speed, and ability to function effectively on edge devices make it highly suitable for deployment in resource-constrained agricultural environments. Moreover, the use of explainability techniques enhances the system’s credibility and practical relevance, especially in settings where stakeholders need to understand and validate AI-generated insights. Real-world validation, conducted across various environmental conditions and regions, demonstrated the model’s adaptability and robustness, confirming its readiness for field deployment. By bridging the gap between advanced AI technologies and real-time field-level agricultural needs, this research contributes meaningfully to the ongoing efforts to modernize disease detection and improve crop health management. It enables timely interventions, reduces dependency on chemical pesticides, and helps optimize yield and resource utilization. In doing so, it supports broader goals of sustainable agriculture, food security, and rural economic development. The system’s open and scalable design allows for extension to other crops and disease types, further increasing its impact and applicability. Continued refinement, integration with mobile platforms, and stakeholder training can ensure its widespread adoption and long-term success in the agricultural domain.

References

[1] Gupta, A., Chug, A., & Singh, A. P. (2024). Potato disease prediction using machine learning, image processing and IoT–a systematic literature survey. Journal of Crop Improvement, 38(2), 95-137. [2] Wisidagama, N. S., Marikar, F. M. M. T., & Sirisuriya, M. (2024). A COMPREHENSIVE REVIEW ON SUITABLE IMAGE PROCESSING AND MACHINE LEARNING TECHNIQUE FOR DISEASE IDENTIFICATION OF TOMATO AND POTATO PLANTS. Automation of Technological & Business Processes/Avtomatizaciâ Tehnologiceskih i Biznes-Processov, 16(1). [3] Reis, H. C., & Turk, V. (2024). Potato leaf disease detection with a novel deep learning model based on depthwise separable convolution and transformer networks. Engineering Applications of Artificial Intelligence, 133, 108307. [4] Pandey, D., Singh, R., Awasthi, A., Bewerwal, A., & Sagar, L. (2024, March). Design and Development of Machine Learning Approaches for Tomato Leaf Disease Identification and Categorization. In 2024 2nd International Conference on Disruptive Technologies (ICDT) (pp. 1325-1330). IEEE. [5] Bak?r, H. (2024). Evaluating the impact of tuned pre-trained architectures\' feature maps on deep learning model performance for tomato disease detection. Multimedia Tools and Applications, 83(6), 18147-18168. [6] YILDIZ, M. B., HAFIF, M. F., KOKSOY, E. K., & KUR?UN, R. (2024). Classification of Diseases in Tomato Leaves Using Deep Learning Methods. Intelligent Methods In Engineering Sciences, 3(1), 22-36. [7] Anitha, P., Kuthkunja, A., Aditya, G., & Anvitha, V. (2024, January). Analysis of Leaf Disease Detection in the Solanaceae family plants using Machine Learning Algorithms. In 2024 International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics (IITCEE) (pp. 1-6). IEEE. [8] Abbas, A., Maqsood, U., Rehman, S. U., Mahmood, K., AlSaedi, T., & Kundi, M. (2024). An Artificial Intelligence Framework for Disease Detection in Potato Plants. Engineering, Technology & Applied Science Research, 14(1), 12628-12635. [9] Sasan, T. K., Monga, J. K., Kaur, J., & Chawla, J. (2024, March). VGG16-PotatoGuard: A Deep Learning Approach to Detecting Leaf Diseases in Potatoes. In 2024 International Conference on Emerging Smart Computing and Informatics (ESCI) (pp. 1-6). IEEE. [10] Lekha, J., Saraswathi, S., Suryaprabha, D., & Thomas, N. (2024, March). Tomato Leaf Disease Detection using Machine Learning Model. In Proceedings of the 1st International Conference on Artificial Intelligence, Communication, IoT, Data Engineering and Security, IACIDS 2023, 23-25 November 2023, Lavasa, Pune, India. [11] Anim-Ayeko, A. O., Schillaci, C., & Lipani, A. (2023). Automatic blight disease detection in potato (Solanum tuberosum L.) and tomato (Solanum lycopersicum, L. 1753) plants using deep learning. Smart Agricultural Technology, 4, 100178. [12] Kumar, Y., Singh, R., Moudgil, M. R., & Kamini. (2023). A systematic review of different categories of plant disease detection using deep learning-based approaches. Archives of Computational Methods in Engineering, 30(8), 4757-4779. [13] Wani, J. A., Sharma, S., Muzamil, M., Ahmed, S., Sharma, S., & Singh, S. (2022). Machine learning and deep learning based computational techniques in automatic agricultural diseases detection: Methodologies, applications, and challenges. Archives of Computational methods in Engineering, 29(1), 641-677. [14] Eligar, V., Patil, U., & Mudenagudi, U. (2022). Performance Analysis of Deep Learning Algorithms Toward Disease Detection: Tomato and Potato Plant as Use-Cases. In Intelligent and Cloud Computing: Proceedings of ICICC 2021 (pp. 595-606). Singapore: Springer Nature Singapore. [15] Chowdhury, M. E., Rahman, T., Khandakar, A., Ibtehaz, N., Khan, A. U., Khan, M. S.,... & Ali, S. H. M. (2021). Tomato leaf diseases detection using deep learning technique. Technology in Agriculture, 453. [16] Rashid, J., Khan, I., Ali, G., Almotiri, S. H., AlGhamdi, M. A., & Masood, K. (2021). Multi-level deep learning model for potato leaf disease recognition. Electronics, 10(17), 2064. [17] Tiwari, D., Ashish, M., Gangwar, N., Sharma, A., Patel, S., & Bhardwaj, S. (2020, May). Potato leaf diseases detection using deep learning. In 2020 4th international conference on intelligent computing and control systems (ICICCS) (pp. 461-466). IEEE. [18] Kothari, D., Mishra, H., Gharat, M., Pandey, V., Gharat, M., & Thakur, R. (2022). Potato leaf disease detection using deep learning. Int. J. Eng. Res. Technol, 11(11). [19] Agarwal, M., Singh, A., Arjaria, S., Sinha, A., & Gupta, S. (2020). ToLeD: Tomato leaf disease detection using convolution neural network. Procedia Computer Science, 167, 293-301. [20] Mahum, R., Munir, H., Mughal, Z. U. N., Awais, M., Sher Khan, F., Saqlain, M., ... & Tlili, I. (2023). A novel framework for potato leaf disease detection using an efficient deep learning model. Human and Ecological Risk Assessment: An International Journal, 29(2), 303-326. [21] Abadi, M. (2016). TensorFlow: learning functions at scale. In Proceedings of the 21st ACM SIGPLAN International Conference on Functional Programming (pp. 1–1). ACM Digital Library. https://doi.org/10.1145/2951913.2976746 [22] Agarwal, M., Singh, A., Arjaria, S., Sinha, A., & Gupta, S. (2020). ToLeD: Tomato leaf disease detection using convolution neural network. Procedia Computer Science, 167, 293–301. https://doi.org/10.1016/j.procs.2020.03.225 [23] Akbar, M., Ullah, M., Shah, B., Khan, R. U., Hussain, T., Ali, F., et al. (2022). An effective deep learning approach for the classification of bacteriosis in peach leaves. Frontiers in Plant Science, 13. https://doi.org/10.3389/fpls.2022.1064854 [24] Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., et al. (2021). Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions. Journal of Big Data, 8, 1–74. https://doi.org/10.1186/s40537-021-00444-8 [25] Anjna,, Sood, M., & Singh, P. K. (2020). Hybrid system for detection and classification of plant disease using qualitative texture features analysis. Procedia Computer Science, 167, 1056–1065. https://doi.org/10.1016/j.procs.2020.03.404 [26] Antonellis, G., Gavras, A. G., Panagiotou, M., Kutter, B. L., Guerrini, G., Sander, A. C., et al. (2015). Shake table test of large-scale bridge columns supported on rocking shallow foundations. Journal of Geotechnical and Geoenvironmental Engineering, 12, 04015009. https://doi.org/10.1061/(ASCE)GT.1943-5606.0001284 [27] Arcaini, P., Bombarda, A., Bonfanti, S., & Gargantini, A. (2020). Dealing with robustness of convolutional neural networks for image classification. In Proceedings - 2020 IEEE International Conference on Artificial Intelligence Testing, AITest 2020 (pp. 7–14). IEEE. https://doi.org/10.1109/AITEST49225.2020.00009 [28] Chen, J., Chen, J., Zhang, D., Sun, Y., & Nanehkaran, Y. A. (2020). Using deep transfer learning for image-based plant disease identification. Computers and Electronics in Agriculture, 173, 105393. https://doi.org/10.1016/j.compag.2020.105393 [29] Brahimi, M., Arsenovic, M., Laraba, S., Sladojevic, S., Boukhalfa, K., & Moussaoui, A. (2018). Deep learning for plant diseases: Detection and saliency map visualization. In Human and Machine Learning: Visible, Explainable, Trustworthy, Transparent (pp. 93–117). https://doi.org/10.1007/978-3-319-90403-0_6 [30] Barbedo, J. G. A. (2018). Factors influencing the use of deep learning for plant disease recognition. Biosystems Engineering, 172, 84–91. https://doi.org/10.1016/j.biosystemseng.2018.05.013 [31] Bahrampour, S., Ramakrishnan, N., Schott, L., & Shah, M. (2015). Comparative study of caffe, neon, theano, and torch for deep learning. arXiv preprint arXiv:1511.06435, 1–9. https://doi.org/10.48550/arXiv.1511.06435 [32] Convolutional encoder-decoder architecture for image segmentation. (2016). IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(12), 2481–2495. https://doi.org/10.1109/TPAMI.2016.2644615 [33] Cedric, L. S., Adoni, W. Y. H., Aworka, R., Zoueu, J. T., Mutombo, F. K., Krichen, M., et al. (2022). Crops yield prediction based on machine learning models: Case of West African countries. Smart Agriculture Technologies, 2(March), 100049. https://doi.org/10.1016/j.atech.2022.100049 [34] Chiu, M. T., Xu, X., Wei, Y., Huang, Z., Schwing, A., Brunner, R., et al. (2020). Agriculture-Vision: A Large Aerial Image Database for Agricultural Pattern Analysis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (pp.2825–2835). IEEE. https://doi.org/10.1109/CVPR42600.2020.00290 [35] Chung, Y., Ahn, S., Yang, J., & Lee, J. (2017). Comparison of deep learning frameworks: About theano, tensorflow, and cognitive toolkit. Journal of Intelligent Information Systems, 23(2), 1–17. https://doi.org/10.13088/jiis.2020.26.4.027 [36] Dechant, C., Wiesner-Hanks, T., Chen, S., Stewart, E. L., Yosinski, J., Gore, M. A., et al. (2017). Automated identification of northern leaf blight-infected maize plants from field imagery using deep learning. Phytopathology, 107, 1426–1432. https://doi.org/10.1094/PHYTO-11-16-0417-R [37] Dhaka, V. S., Meena, S. V., Rani, G., Sinwar, D., Ijaz, M. F., & Woz?niak, M. (2021). A survey of deep convolutional neural networks applied for prediction of plant leaf diseases. Sensors, 21(14), 4749. https://doi.org/10.3390/s21144749 [38] Dillon, J. V., Langmore, I., Tran, D., Brevdo, E., Vasudevan, S., Moore, D., et al. (2017). Tensorflow distributions. arXiv preprint arXiv:1711.10604, 1–10. https://doi.org/10.48550/arXiv.1711.10604 [39] Domingues, T., Brandão, T., & Ferreira, J. C. (2022). Machine learning for detection and prediction of crop diseases and pests: A comprehensive survey. Agriculture, 12(9), 1350. https://doi.org/10.3390/agriculture12091350 [40] Drenkow, N., Sani, N., Shpitser, I., & Unberath, M. (2021). Robustness in deep learning for computer vision: Mind the gap? arXiv preprint arXiv:2112.00639, 1–23. https://doi.org/10.48550/arXiv.2112.00639

Copyright

Copyright © 2025 Amit Banga, Dr. Anil Dudi. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.