|
Chidinma, E. U.
Department of Computer Engineering Michael Okpara University of Agriculture Umudike, Abia State Nigeria.
Chiagunye, T. T.
Department of Computer Engineering Michael Okpara University of Agriculture Umudike, Abia State Nigeria.
Ilo, S. F.
Department of Computer Engineering Michael Okpara University of Agriculture Umudike, Abia State Nigeria.
Nwali, U. E.
Department of Computer Engineering Michael Okpara University of Agriculture Umudike, Abia State Nigeria.
ABSTRACT
This research focuses on
the development of a Convolutional Neural Network (CNN)-based system for the
early detection and classification of tomato plant diseases, leveraging
transfer learning with the InceptionV3 architecture. The system addresses
critical challenges faced by smallholder farmers in Nigeria, such as economic
losses and reduced yields due to diseases like Early Blight, Late Blight, and
Bacterial Spot. By utilizing a dataset from PlantVillage augmented with
field-collected images, the system achieved a validation accuracy of 91.50%, demonstrating robust performance in
disease identification. The research employs a web-based interface for
real-time image uploads and camera integration, ensuring accessibility for
farmers with limited technical expertise. An integrated AI chat assistant
provides actionable treatment recommendations, bridging the gap between
diagnosis and practical solutions. The system's design follows Agile Scrum
methodology, incorporating data pre-processing, model training, and performance
evaluation phases. Key findings highlight the model's effectiveness in reducing
crop losses and promoting sustainable agriculture. However, limitations such as
data variability and internet dependency suggest future improvements, including
offline mobile deployment and expanded datasets. This study contributes to
precision agriculture by offering a scalable, user-friendly tool for enhancing
food security and farmer livelihoods in resource-constrained regions.
Keywords: Tomato plant diseases, Convolutional Neural Network (CNN), InceptionV3, transfer learning, precision agriculture, disease detection, sustainable farming.
https://doi.org/10.33922/j.ujet_v11i2_3
|
View: 19 | Download: 16
Published
Monday, July 14, 2025
Issue
Vol. 11 No. 2, December 2025
Article Section
GENERAL
The contents of the articles are the sole opinion of the author(s) and not of UJET.
|
