AgroBot: AI-based Leaf Disease Detection with Intelligent Chatbot

Agriculture plays a vital role in global food production; however, plant diseases significantly reduce crop yield and quality, posing a major challenge for farmers. Early detection of leaf diseases is difficult due to limited access to expert knowledge and timely diagnosis. To address this problem, this paper proposes AgroBot, an AI-driven leaf-disease detection system integrated with an intelligent chatbot to provide smart agricultural assistance. The system allows users to upload leaf images through a web-based interface, which are then processed using Deep Learning techniques. A ResNet model, specifically based on the Residual Network architecture, is employed for accurate disease classification. To mitigate dataset imbalance and improve model generalization, a Generative Adversarial Network (GAN) is used to augment the data by generating realistic synthetic leaf images. The model is trained and evaluated on the PlantVillage dataset, consisting of diverse crop leaf images under varying conditions. Experimental results demonstrate that the proposed system achieves an accuracy of approximately 95–98% in disease classification, showing robustness even under noisy and imbalanced data scenarios. In addition to detecting diseases, the system provides severity analysis, along with tailored treatment recommendations and preventive measures. Furthermore, AgroBot integrates an intelligent chatbot that delivers real-time, context-aware guidance to farmers, assisting them in understanding disease symptoms, suggesting remedies, and recommending relevant government agricultural schemes and support services. This enhances accessibility to critical agricultural knowledge and promotes informed decisionmaking. The proposed system demonstrates significant potential in advancing precision agriculture, improving crop health management, and supporting farmers with scalable, technology-driven solutions.