Research Article
AI-Powered Corn Disease Classification Using Deep Transfer Learning
@INPROCEEDINGS{10.1007/978-3-031-86493-3_28, author={Moussa Mahamat Boukar and Assia Aboubakar Mahamat and Hassane Hamdan and Usman Abubakar Bello}, title={AI-Powered Corn Disease Classification Using Deep Transfer Learning}, proceedings={Innovations and Interdisciplinary Solutions for Underserved Areas. 7th International Conference, InterSol 2024, Dakar, Senegal, July 3--4, 2024, Proceedings}, proceedings_a={INTERSOL}, year={2025}, month={4}, keywords={AI Corn disease Identification Deep Transfer-Learning Machine Learning}, doi={10.1007/978-3-031-86493-3_28} }- Moussa Mahamat Boukar
Assia Aboubakar Mahamat
Hassane Hamdan
Usman Abubakar Bello
Year: 2025
AI-Powered Corn Disease Classification Using Deep Transfer Learning
INTERSOL
Springer
DOI: 10.1007/978-3-031-86493-3_28
Abstract
Corn, a vital agricultural crop, and essential food source, plays a crucial role in the global food chain and serves as a raw material for various industrial applications like biofuels. Small-scale corn cultivation sustains livelihoods in developing nations, but these crops are highly susceptible to diseases. Extreme weather conditions can exacerbate these diseases, leading to significant declines in agricultural yields.
Advancements in artificial intelligence (AI), particularly deep learning algorithms, offer promising solutions. This study explores the application of deep transfer learning for classifying three distinct corn leaf conditions: rust, northern leaf blight, and healthy plants. By utilizing corn leaf images as input and leveraging convolutional neural networks, the proposed approach eliminates the need for complex pre-processing or manual feature extraction.
Employing well-established deep learning models (VGG19, GoogleNet, and ResNet50) and rigorous evaluation methods with various data splitting scenarios, the study achieved remarkable mean accuracies of 96%, 99%, and 75% in distinguishing the three classes. These results demonstrate the potential for developing practical applications to assist farmers and plant pathologists in accurately and swiftly identifying corn diseases, enabling them to implement appropriate treatment measures.
