An Integrated Diagnostic Model for Identifying and Diagnosing Groundnut Leaf Diseases
DOI:
https://doi.org/10.64290/bima.v9i2B.1295Keywords:
Image Processing Convolutional Neural Networks (CNN’s), Residual Network (ResNet50), Visual Geometry Group(VGG16), Bayesian Optimization (BO), Computer Vision.Abstract
Groundnut (Arachis hypogaea L.), also known as peanut, serves as a significant source of edible oil and protein, making it essential for the agricultural economy and food security. However, groundnut plants are susceptible to various diseases, notably leaf diseases. Early and accurate identification of these diseases is crucial for implementing timely management strategies to mitigate losses. In recent years, advancements in deep learning especially convolutional neural networks (CNNs), have revolutionized image recognition tasks, including plant disease identification. This study employed the use of ResNet50 and VGG16 Convolutional Neural Networks (CNNs) with Bayesian Optimization (BO). The dataset comprises both infected and uninfected groundnut leaf images, categorized into six folders based on their status. Performance evaluation demonstrates that ResNet50 and VGG16 model’s ability to accurately identify groundnut leaves diseases. Results indicate that the hybrid models ResNet50 and VGG16 effectively captures and distinguishes between infected and uninfected groundnut leaves disease with a high degree of accuracy of 98.72% and 98.22% respectively.
