ML Approaches for Real-Time Cotton Leaf Disease Detection and ‎Severity Prediction Through Multi-Modal Data Integration and ‎Transfer Learning

  • Authors

    • M. DhanaLakshmi Computer Science & Engineering GIET University
    • Bidush Kumar Sahoo Computer Science & Engineering GIET University
    • Rajendra Kumar Ganiya Professor, ‎Department of Computer Science ‎and Engineering,‎ Koneru Lakshmaiah Education ‎ Foundation, Vaddeswaram, ‎Guntur, A.P., 522302, India
    https://doi.org/10.14419/nmkdgd22

    Received date: July 13, 2025

    Accepted date: July 25, 2025

    Published date: November 1, 2025

  • Plant Disease Detection; Convolutional Neural Networks (CNNs); DL In Agriculture; Transfer ‎Learning; Multi-Modal Data Integration; Precision Agriculture; Crop Disease Classification; ‎Image-Based Disease Diagnosis; Data Augmentation; Real-Time Disease Monitoring

  • Abstract

    This research mainly focused on with the exploration aming deep learning and utilized ‎through a CNN model, Transfer learning, and integrated with multi-model identification of plant ‎diseases. The main objective is to identify the agronomic flaws in conventional farming ‎practices that led to wasteful and unproductive approaches, including the failure to diagnose ‎malaria, a major health risk that reduced agricultural yields. Using the Plant Village dataset, ‎which includes 54,000 images of plant leaves annotated with disease kind, we were able to ‎construct a better field readiness model by identifying 38 different illnesses using a subset of ‎these photos that provides around 30,000 normal and diseased leaves. We tried a number of ‎different ways to add to the data so that performance would go up and overfitting would be ‎avoided. Accuracy, precision, and recall for illness classes are all above average in our CNN-based model, which achieves 94.9% on average. The confusion matrix revealed only a few ‎misclassified photos, with the best accuracy for healthy leaves (98.0%), followed by "Tomato ‎Early Blight" (94.9%) and "Potato Late Blight" (96.2%). Refined a pretrained EfficientNet ‎model to achieve 96.5% accuracy. Madonna and others. This model can also train up to 25% ‎faster than other designs like ResNet, which makes it good for real-time mobile and edge ‎applications. When the environmental data (temperature, humidity) and pictures of leaves were ‎put together, the accuracy was similarly 97.1%. These approaches worked well together to find ‎illness heterogeneity caused by the environment. This work demonstrates that deep learning ‎‎(DL) may be efficiently employed for early disease detection in plantations, a fundamental ‎aspect of precision agriculture‎.

  • References

    1. Iftikhar, Malik Faisal, et al. "Plant Disease Management: A Fine-Tuned Enhanced CNN Approach with Mobile App Integration for Early Detection and Classification." Artificial Intelligence Review, 2024. https://doi.org/10.21203/rs.3.rs-3782911/v1.
    2. Nichat, Mangesh K., and S. E. Yedey. "DL Techniques for Identification of Different Malvaceae Plant Leaf Diseases." EAI Endorsed Transactions on Internet of Things, 2024. https://doi.org/10.4108/eetiot.5394.
    3. Abouhawwash, Mohamed, and Nihal N. Mostafa. "DL for Coffee Leaf Diseases Detection in Precision Agriculture." Optimization in Agriculture 1 (2024): 129-136. https://doi.org/10.61356/j.oia.2024.1292.
    4. Baig, Muhammad Daniyal, et al. "Transfer Learning Approaches for Leaf Diseased Portion Detection." Indian Scientific Journal of Research in En-gineering and Management, 2023.
    5. Baig, Muhammad Daniyal, et al. "Leaf Diseases Detection Empowered with Transfer Learning Model." [Please provide the correct journal name]
    6. Baig, Muhammad Daniyal, et al. "Estimation of Cotton Leaf Infection Using DL VCG16 Model on Segmented Cotton Leaf Images Contained through Deformable Model." Neuroquantology, 2023.
    7. Amin, Javeria, et al. "Explainable Neural Network for Classification of Cotton Leaf Diseases." Agriculture, 2022. https://doi.org/10.3390/agriculture12122029.
    8. Memon, Muhammad Suleman, et al. "Identification of Leaf Diseases of Different Crops Using Modified ResNet50." [Please provide the correct journal name]
    9. Nazeer, Rubaina, et al. "Detection of Cotton Leaf Curl Disease’s Susceptibility Scale Level Based on DL." [Please provide the correct journal name]
    10. Gülmez, Burak. "A Novel DL Model with the Grey Wolf Optimization Algorithm for Cotton Disease Detection." Journal of Universal Computer Science, 2023. https://doi.org/10.3897/jucs.94183.
    11. Baig, Muhammad Daniyal, et al. "Transfer Learning Approaches for Leaf Diseased Portion Detection." Indian Scientific Journal of Research in En-gineering and Management, 2023.
    12. Baig, Muhammad Daniyal, et al. "Leaf Diseases Detection Empowered with Transfer Learning Model." [Please provide the correct journal name]
    13. Baig, Muhammad Daniyal, et al. "Estimation of Cotton Leaf Infection Using DL VCG16 Model on Segmented Cotton Leaf Images Contained through Deformable Model." Neuroquantology, 2023.
    14. Chen, L., Li, S., Bai, Q., Yang, J., Jiang, S., & Miao, Y. "Review of Image Classification Algorithms Based on Convolutional Neural Networks." Remote Sensing, 13(22), 4712. https://doi.org/10.3390/rs13224712.
    15. Gülmez, B. (2021). "Prediction of retail prices of roasted coffee by time series analysis." In International Anatolian Congress on Coffee and Cocoa (p. 63). Malatya, Turkey.
    16. Gülmez, B. (2022). "A novel deep neural network model based Xception and genetic algorithm for detection of COVID-19 from X-ray images." Annals of Operations Research. https://doi.org/10.1007/s10479-022-05151-y.
    17. Gülmez, B. (2023a). "İşyerigüvenliğiiçinderinöğrenmeilebarettakılmasıtespiti." In International Conference on Engineering and Applied Natural Sci-ences, Konya, Türkiye, 14-17 Jan 2023.
    18. Kumar, S., Jain, A., Shukla, A. P., Singh, S., Raja, R., Rani, S., et al. "A Comparative Analysis of Machine Learning Algorithms for Detection of Organic and Nonorganic Cotton Diseases." Mathematical Problems in Engineering, 2021, e1790171. https://doi.org/10.1155/2021/1790171.
    19. Li, Y., & Yang, J. (2020). "Few-shot cotton pest recognition and terminal realization." Computers and Electronics in Agriculture, 169, 105240. https://doi.org/10.1016/j.compag.2020.105240.
    20. Liu, Y., Sun, P., Wergeles, N., & Shang, Y. "A survey and performance evaluation of DL methods for small object detection." Expert Systems with Applications, 172, 114602. https://doi.org/10.1016/j.eswa.2021.114602.
    21. Memon, M. S., Kumar, P., & Iqbal, R. (2022). "Meta Deep Learn Leaf Disease Identification Model for Cotton Crop." Computers, 11(7), 102. https://doi.org/10.3390/computers11070102.
    22. Mirjalili, S., Mirjalili, S. M., & Lewis, A. "Grey Wolf Optimizer." Advances in Engineering Software, 69, 46–61. https://doi.org/10.1016/j.advengsoft.2013.12.007.
    23. Momeny, M., Latif, A. M., Agha Sarram, M., Sheikhpour, R., & Zhang, Y. D. (2021). "A noise robust convolutional neural network for image clas-sification." Results in Engineering, 10, 100225. https://doi.org/10.1016/j.rineng.2021.100225.
    24. Noon, S. K., Amjad, M., Ali Qureshi, M., & Mannan, A. "Computationally light DL framework to recognize cotton leaf diseases." Journal of Intel-ligent & Fuzzy Systems, 40(6), 12383–12398. https://doi.org/10.3233/JIFS-210516.
    25. Patil, B. V., & Patil, P. S. "Computational Method for Cotton Plant Disease Detection of Crop Management Using DL and Internet of Things Plat-forms." In V. Suma, N. Bouhmala& H. Wang (Eds.), Evolutionary Computing and Mobile Sustainable Networks (pp. 875–885). Singapore: Spring-er. https://doi.org/10.1007/978-981-15-5258-8_81.
    26. Prashar, K., Talwar, R., & Kant, C. "CNN based on Overlapping Pooling Method and Multi-layered Learning with SVM & KNN for American Cotton Leaf Disease Recognition." In 2019 International Conference on Automation, Computational and Technology Management (ICACTM) (pp. 330–333). London, UK. https://doi.org/10.1109/ICACTM.2019.8776730.
    27. Savran Kızıltepe, R., Gan, J. Q., & Escobar, J. J. "A novel keyframe extraction method for video classification using deep neural networks." Neural Computing and Applications. https://doi.org/10.1007/s00521-021-06322-x.
    28. Sivakumar, P., Mohan, N. S. R., Kavya, P., & Teja, P. V. S. "Leaf Disease Identification: Enhanced Cotton Leaf Disease Identification Using Deep CNN Models." In 2021 IEEE International Conference on Intelligent Systems, Smart and Green Technologies (ICISSGT) (pp. 22–26). Visakhapat-nam, India. https://doi.org/10.1109/ICISSGT52025.2021.00016.
    29. Sohrab, S. S., Azhar, E. I., Kamal, M. A., Bhattacharya, P. S., & Rana, D. "Genetic variability of Cotton leaf curl betasatellite in Northern India." Saudi Journal of Biological Sciences, 21(6), 626–631. https://doi.org/10.1016/j.sjbs.2014.11.006.

  • Downloads

  • How to Cite

    DhanaLakshmi, M., Sahoo, B. K. ., & Ganiya, R. K. . (2025). ML Approaches for Real-Time Cotton Leaf Disease Detection and ‎Severity Prediction Through Multi-Modal Data Integration and ‎Transfer Learning. International Journal of Basic and Applied Sciences, 14(SI-1), 519-528. https://doi.org/10.14419/nmkdgd22