AgriAid: AI-Powered Agricultural Assistant for Inclusive Plant Disease Detection

Abstract

Agriculture remains the backbone of the economy, yet farmers frequently grapple with yield- limiting challenges such as unidentified plant diseases, unpredictable weather patterns, and volatile marketprices.Whilerecentadvancementsindeeplearninghavedemonstratedhighaccuracyindisease classification under laboratory conditions [1], many existing solutions fail to address real-world constraints such as linguistic barriers, low literacy, and intermittent internet connectivity. This paper presents AgriAid, an inclusive, AI-powered mobile ecosystem designed to democratize access to agricultural expertise. The proposed system employs a multimodal input framework accepting voice, text, and image data to facilitate intuitive interaction for users across diverse demographics. Leveraging Convolutional Neural Networks (CNN) for high-precision image-based disease detection and the Google Gemini API for generative, context-aware agricultural advisory, the platform provides real-time diagnosis, remedial measures, and personalized crop recommendations. Furthermore, the application integrates hyper local 7-day weather forecasts, government scheme eligibility checks, and live market price tracking into a multilingual interface. Developed using React Native and Node.js, the system is optimized for offline capability, ensuring robust decision support even in remote, low-connectivity environments.