An Advanced CNN-Driven Image Analysis System for Proactive Prediction and Automated Detection Drivers Disease Conditions in Rice Crop Cultivation

Abstract

This work proposes a CNN-driven image analysis framework for the automated identification and classification of rice crop diseases using deep learning within a web-enabled environment. The system utilizes a rice leaf image dataset consisting of both healthy samples and diseased leaves affected by bacterial leaf blight, brown spot, and leaf smut. To improve data quality and model reliability, preprocessing operations such as image resizing, normalization, and augmentation are performed before training. A Convolutional Neural Network (CNN) is employed to automatically learn discriminative features from leaf images and perform disease classification without requiring manual feature extraction. The trained model is deployed through a Flask-based web application that supports secure user authentication, image uploading, prediction generation, and real-time disease diagnosis. In addition, the framework provides appropriate treatment recommendations based on the detected disease category. Performance evaluation is carried out using visualization tools, including accuracy curves and training-performance graphs, to analyze learning behavior and classification effectiveness. The proposed solution offers a scalable, accurate, and efficient approach for automated rice disease detection, thereby supporting precision farming practices and enhancing agricultural productivity