Rice Leaf Disease Detection System Using the Convolutional Neural Network (ResNet)

Rice is one of the most widely produced crops in the world and leaf diseases are hampering the productivity of this crop. To avoid the spread of these leaf diseases and reduce crop losses, the disease should be diagnosed as early as possible. Recently, Convolutional Neural Networks (CNNs) have been used to detect plant diseases with high accuracy and image classification. This work proposes a method for analyzing rice images and detecting various diseases of rice leaves, such as bacterial leaf blight, brown spot, hispa, leaf blast, leaf scald, leaf streak, narrow brown spot, sheath blight, and tungro, as well as healthy rice leaves. We use a Convolutional Neural Network (CNN) model built on the ResNet architecture, ensuring professional and accurate results. ResNet is highly regarded for its advanced deep learning capabilities and its effectiveness in tackling invisible gradient problems. It offers a reliable and robust framework specifically designed for image classification tasks. This architecture provides an effective and robust solution to the image classification problems, thus providing reliable results. Image analysis was required to classify rice diseases and detect the presence of diseases on leaves and manage them efficiently. The dataset selected for this analysis consisted of a total of 1000 rice leaf images, which included both healthy and disease-affected leaves. The model performed with excellent accuracy as it detected rice leaf diseases with 95.6% accuracy, making it a potential solution for early detection and control of rice crop diseases