Early Stage Identification of Tomato Leaf Diseases using VGG16 and MobileNet Convolutional Neural Networks

Abstract

This study presents an innovative approach for early identification of tomato leaf diseases, using a combination of VGG16 and MobileNet Convolutional Neural Networks (CNNs). Unlike traditional manual observation methods, this advanced approach offers improved scalability, speed, and accuracy in early disease detection. By integrating these powerful CNN architectures, VGG16 and MobileNet, we achieve an impressive accuracy of approximately 94%, with precision and recall rates of 93% and 92% respectively. These results mark a significant advancement in agricultural technology and plant pathology. Beyond academia, this integrated model offers practical, scalable solutions adaptable to various agricultural settings, potentially revolutionizing crop management and aiding agricultural sustainability amidst environmental and climatic challenges. This research demonstrates the potential of integrating cutting-edge technologies to address longstanding agricultural challenges and inspire future innovations in the field.