Authors: Sukanya, G.Bharath Kumar, Karthik.D, Nikhil Reddy, ChannaKeshwa
Abstract: Crop diseases pose a major global threat to agricultural productivity, farmer income, and overall food security. Widespread disease outbreaks reduce crop quality, decrease yield, and contribute to economic instability—especially in regions dependent on agriculture for livelihood. The complexity of crop diseases arises from diverse environmental conditions, varying plant species, and the presence of multiple visually similar infections. Addressing these challenges requires a systematic, data-driven approach capable of identifying hidden patterns and supporting farmers and agricultural experts with timely, actionable insights. This project presents the design and development of an AI- Driven Crop Disease Prediction and Monitoring Dashboard, an interactive platform built using Streamlit. The dashboard enables visualization, prediction, and analysis of plant disease data using a trained Convolutional Neural Network (CNN). The system architecture is organized into three primary layers: the Presentation Layer, the Logic Layer, and the Data Layer. The Presentation Layer provides a user-friendly web interface developed in Streamlit, integrating dynamic components such as real-time prediction panels, probability bars, and comparative disease charts generated with Plotly Express. It also includes essential UI elements such as an image upload section and model output visualization to ensure smooth user interaction. The Logic Layer performs core analytical and computational tasks. It preprocesses leaf images, applies the CNN model for classification, generates confidence scores, and provides diseasespecific treatment recommendations. Pandas handles metadata processing, while session-state management ensures efficient handling of user inputs and outputs. The Data Layer consists of a structured plant disease dataset derived from sources such as PlantVillage, supplemented with augmented images to improve model robustness across lighting and environmental variations.
