AgriVeda AI: An Intelligent Smart Agriculture Platform Using Machine Learning and Explainable AI

This paper presented AgriVeda AI, a comprehensive smart agriculture platform integrating ten AI-powered modules within a unified Flask-React web application. The system delivers strong technical performance: 98.2% crop recommendation accuracy, R-squared of 0.98 for yield prediction, 94% CNN disease detection accuracy, and real-time integration with Open-Meteo and AGMARKNET APIs. The inclusion of SHAP explainability for yield predictions and Grad-CAM visualization for disease detection significantly enhances the transparency and interpretability of AI outputs — critical requirements for farmer adoption and trust. The cross-module integration demonstrated through the AI Shopping List — which automatically consolidates outputs from disease detection, yield prediction, irrigation planning, and crop recommendation — represents a novel contribution to holistic precision agriculture platforms. The modular Flask-React architecture ensures independent maintainability and scalability of each component. Future work will focus on: (1) multilingual support (Kannada, Hindi, Tamil) for broader rural accessibility; (2) mobile application development with offline-capable model inference; (3) IoT sensor integration for real-time soil NPK and moisture data input; (4) expansion of the disease detection model to additional crop species and regional disease variants; (5) time-series yield forecasting incorporating historical farm data; and (6) federated learning for privacy-preserving model improvement across distributed farm data. The AgriVeda AI platform demonstrates the practical feasibility of integrated AI-powered smart agriculture and represents a significant contribution toward digitally empowered Indian farming.

[1]        Ministry of Agriculture & Farmers Welfare, Government of India, "Agriculture Census 2015-16," New Delhi, India, 2019. [2]        S. P. Mohanty, D. P. Hughes, and M. Salathe, "Using deep learning for image-based plant disease detection," Front. Plant Sci., vol. 7, p. 1419, Sep. 2016. [3]        K. P. Ferentinos, "Deep learning models for plant disease detection and diagnosis," Comput. Electron. Agric., vol. 145, pp. 311-318, Feb. 2018. [4]        R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, "Grad-CAM: Visual explanations from deep networks via gradient-based localization," in Proc. IEEE ICCV, pp. 618-626, 2017. [5]        J. H. Jeong, J. P. Resop, N. D. Mueller, D. H. Fleisher, K. Yun, E. E. Butler, D. J. Timlin, K.-J. Shim, J. S. Gerber, V. R. Reddy, and D. Kim, "Random forests for global and regional crop yield predictions," PLoS ONE, vol. 11, no. 6, p. e0156571, 2016. [6]        X. E. Pantazi, D. Moshou, T. Alexandridis, R. L. Whetton, and A. M. Mouazen, "Wheat yield prediction using machine learning and advanced sensing techniques," Comput. Electron. Agric., vol. 121, pp. 57-65, Feb. 2016. [7]        S. M. Lundberg and S.-I. Lee, "A unified approach to interpreting model predictions," in Proc. NIPS, vol. 30, pp. 4765-4774, 2017. [8]        P. Lewis, E. Perez, A. Piktus, F. Petroni, V. Karpukhin, N. Goyal et al., "Retrieval-augmented generation for knowledge-intensive NLP tasks," in Proc. NeurIPS, vol. 33, pp. 9459-9474, 2020. [9]        A. Kamilaris and F. X. Prenafeta-Boldu, "Deep learning in agriculture: A survey," Comput. Electron. Agric., vol. 147, pp. 70-90, Apr. 2018. [10]     D. P. Hughes and M. Salathe, "An open access repository of images on plant health to enable the development of mobile disease diagnostics," arXiv:1511.08060, 2015.