DEEP LEARNING BASED MAIZE PLANT DISEASE IDENTIFICATION SYSTEM

DEEP LEARNING BASED MAIZE PLANT DISEASE IDENTIFICATION SYSTEM

Meghana U1, Mrs. Roopa R2

[1]Student, Department of MCA, BIET, Davangere

[2]Assistant Professor, Department of MCA, BIET, Davangere

ABSTRACT

The agricultural sector faces significant challenges due to plant diseases, which can drastically reduce crop yields and quality. This paper presents a deep learning-based system for the identification of maize plant diseases, leveraging Convolutional Neural Networks (CNNs) to analyze images of maize leaves. The proposed system involves the collection of a comprehensive dataset of maize leaf images, encompassing various disease conditions as well as healthy leaves. These images are preprocessed and augmented to enhance the dataset's quality and diversity. The CNN model is then trained and evaluated using this dataset, achieving high accuracy and robust performance metrics such as precision, recall, and F1-score. Field tests demonstrate the system's practical applicability, with a mobile application enabling farmers to capture images of maize leaves and receive instant diagnostic results. The results indicate that the deep learning-based system is a highly effective tool for automated disease detection, offering significant potential for improving crop management and productivity. Future research directions include expanding the dataset, refining the model architecture, and integrating real-time monitoring capabilities through IoT devices. This system represents a significant advancement in agricultural technology, providing a scalable, accurate, and accessible solution for maize plant disease management.

Keywords: Maize plant diseases, Deep learning, Convolutional Neural Networks (CNNs), Image classification, Disease identification.