Date of Publication :18th April 2024

Abstract:Plant diseases pose substantial challenges to agricultural output, needing early identification and intervention efforts. This paper offers a lightweight Convolutional Neural Network (CNN)-based model for disease classification in rice, wheat, and maize plants, which is implemented in MATLAB R2021a. The image dataset includes both damaged and healthy leaves from the three different crops such as Rice, Wheat, and Corn. The proposed CNN architecture is intended to be both efficient and effective, with convolutional layers, batch normalization, and pooling layers. A split dataset is used for training and evaluation, and real-time disease classification is presented using leaf images provided by the user. Accuracy, precision, recall, and F1 score are performance indicators that demonstrate the model's ability to detect and identify diseases across diverse crop kinds. This unified strategy provides a viable option for automated plant disease control, which advances precision. This method not only provides effective outputs but also better than many states of art methods.

Reference :