Abstract:

ABSTRACT-In recent years, plant leaf diseases have become a significant issue, particularly for important crops like beans, which are a major source of protein worldwide. Diseases such as angular leaf spot and bean rust can significantly impact bean production. Early and accurate identification of these diseases is essential for effective intervention. To address this, a deep learning approach using MobileNetV2 architecture is proposed to classify bean leaf diseases using a public dataset of leaf images. The study focuses on the development of an efficient classification system for identifying bean leaf diseases. It employs MobileNet, a lightweight deep learning model, leveraging the open-source TensorFlow library. The model was trained and tested on a dataset consisting of images of bean leaves, categorized into three classes: two unhealthy classes (angular leaf spot disease and bean rust disease) and one healthy class.The model’s performance was evaluated by training on 1296 images, and the results were compared across different network architectures to determine the most effective configuration. The findings showed that the mobile network developed high classification accuracy, achieving over 97% accuracy on the training dataset** and **92% on the test dataset.

Description:

.