Authors : Avala Sai Lokesh; Dr. R. Sathya Janaki

Volume/Issue : Volume 11 - 2026, Issue 7 - July

Google Scholar : https://tinyurl.com/2smx9pet

Scribd : https://tinyurl.com/2cy7ea6n

DOI : https://doi.org/10.38124/ijisrt/26jul1092

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Agriculture is one of the most important sectors supporting food security, employment, and economic growth worldwide. In India, jackfruit (Artocarpus heterophyllus) is a commercially significant tropical fruit crop cultivated extensively in states such as Kerala, Karnataka, Assam, and Tamil Nadu. Among these regions, Panruti in Cuddalore District, Tamil Nadu, is internationally recognized for producing high-quality jackfruit and received the Geographical Indication (GI) tag due to its superior fruit quality and commercial value. Despite its economic importance, jackfruit cultivation is severely affected by fungal diseases, bacterial infections, and insect pests, leading to considerable reductions in yield, fruit quality, and farmer income. Conventional disease diagnosis depends on visual inspection by experienced farmers or agricultural experts. Such manual diagnosis is often subjective, time-consuming, and inaccurate because many diseases exhibit similar symptoms during their early stages. Furthermore, limited access to agricultural specialists in rural areas delays disease identification and treatment, resulting in excessive pesticide application, increased cultivation costs, and environmental degradation. Recent advances in Artificial Intelligence (AI), Deep Learning (DL), and Computer Vision have enabled the development of intelligent plant disease detection systems capable of recognizing complex disease patterns directly from digital images. However, many existing systems rely solely on Convolutional Neural Networks (CNNs), which may have limited ability to capture global contextual information. Moreover, most deep learning models function as "black-box" systems, preventing users from understanding how predictions are generated. To address these limitations, this research proposes AgriVisionXAI, an Explainable Artificial Intelligence (XAI)- based framework for early detection of jackfruit diseases and intelligent recommendation of organic treatments using Hybrid Deep Learning Models. The proposed framework integrates EfficientNet-B3 for efficient feature extraction, Vision Transformer (ViT) for global contextual learning, and Gradient-weighted Class Activation Mapping (Grad-CAM) together with SHapley Additive exPlanations (SHAP) to generate interpretable visual explanations. The hybrid architecture improves disease classification accuracy while enhancing transparency and user trust. The proposed system utilizes a comprehensive image dataset consisting of healthy and diseased jackfruit leaves, fruits, stems, and bark collected from orchards in Panruti and other agricultural regions. The preprocessing pipeline includes image resizing, noise removal, contrast enhancement, histogram equalization, data augmentation, and normalization to improve model robustness under varying environmental conditions. The classification module identifies multiple diseases including Anthracnose, Leaf Spot, Pink Disease, Rhizopus Fruit Rot, Stem Borer Damage, and Fruit Rot. Following disease prediction, an intelligent Organic Recommendation Engine automatically provides scientifically validated eco-friendly treatment suggestions such as Panchagavya, Neem Oil, Trichoderma viride, Pseudomonas fluorescens, Bordeaux mixture, compost tea, and other biological formulations. The recommendation module also includes dosage, preparation procedure, spraying intervals, and preventive management practices. The performance of the proposed framework is evaluated using Accuracy, Precision, Recall, F1-score, Specificity, ROC-AUC Score, Matthews Correlation Coefficient (MCC), Cohen's Kappa, and Confusion Matrix analysis. Prototype evaluation indicates that the proposed hybrid architecture can achieve higher classification accuracy than conventional CNN-based systems while providing interpretable predictions suitable for practical agricultural deployment. The proposed framework supports sustainable precision agriculture by reducing disease diagnosis time, minimizing unnecessary pesticide usage, preserving the export quality of GI-tagged Panruti jackfruit, and promoting environmentally friendly farming practices. The system can further be extended to mobile applications, edge AI devices, unmanned aerial vehicles (UAVs), and Internet of Things (IoT)-based smart farming platforms for real-time disease monitoring and decision support.

Keywords : Artificial Intelligence, Explainable Artificial Intelligence, Deep Learning, Computer Vision, EfficientNet-B3,Vision Transformer,Grad-CAM,SHAP,Jackfruit Disease Detection, Organic Farming, Precision Agriculture, Image Classification, Plant Disease Detection, Sustainable Agriculture, Machine Learning.

References :

1. J. Deng, W. Dong, R. Socher, L. J. Li, K. Li, and L. Fei-Fei, "ImageNet: A Large-Scale Hierarchical Image Database," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Miami, FL, USA, pp. 248–255, 2009.
2. M. Tan and Q. Le, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks," Proceedings of the 36th International Conference on Machine Learning (ICML), pp. 6105–6114, 2019.
3. A. Dosovitskiy et al., "An Image is Worth 16×16 Words: Transformers for Image Recognition at Scale," International Conference on Learning Representations (ICLR), 2021.
4. R. R. Selvaraju et al., "Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization," Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 618–626, 2017.
5. S. M. Lundberg and S. I. Lee, "A Unified Approach to Interpreting Model Predictions," Advances in Neural Information Processing Systems (NeurIPS), vol. 30, 2017.
6. K. He, X. Zhang, S. Ren, and J. Sun, "Deep Residual Learning for Image Recognition," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778, 2016.
7. G. Huang, Z. Liu, L. van der Maaten, and K. Weinberger, "Densely Connected Convolutional Networks," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4700–4708, 2017.
8. M. Sandler et al., "MobileNetV2: Inverted Residuals and Linear Bottlenecks," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4510–4520, 2018.
9. A. Howard et al., "Searching for MobileNetV3," Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 1314–1324, 2019.
10. D. P. Kingma and J. Ba, "Adam: A Method for Stochastic Optimization," International Conference on Learning Representations (ICLR), 2015.
11. I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA, USA: MIT Press, 2016.
12. C. M. Bishop, Pattern Recognition and Machine Learning. Springer, 2006.
13. S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, 4th ed. Pearson, 2021.
14. G. James, D. Witten, T. Hastie, and R. Tibshirani, An Introduction to Statistical Learning, 2nd ed. Springer, 2021.
15. T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, 2nd ed. Springer, 2009.
16. R. C. Gonzalez and R. E. Woods, Digital Image Processing, 4th ed. Pearson, 2018.
17. R. Szeliski, Computer Vision: Algorithms and Applications, 2nd ed. Springer, 2022.
18. Food and Agriculture Organization (FAO), The State of Food and Agriculture. Rome, Italy.
19. Indian Council of Agricultural Research (ICAR), Horticultural Statistics at a Glance. New Delhi, India.
20. Tamil Nadu Agricultural University (TNAU), Crop Production Guide – Jackfruit. Coimbatore, Tamil Nadu.
21. PlantVillage, "Open Access Plant Disease Dataset."
22. Kaggle, "Plant Disease Image Dataset."
23. World Bank, "Agriculture and Rural Development Data."
24. Y. LeCun, Y. Bengio, and G. Hinton, "Deep Learning," Nature, vol. 521, no. 7553, pp. 436–444, 2015.
25. A. Krizhevsky, I. Sutskever, and G. E. Hinton, "ImageNet Classification with Deep Convolutional Neural Networks," Communications of the ACM, vol. 60, no. 6, pp. 84–90, 2017.