This study presents a hybrid deep learning approach that combines Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) for maize leaf disease detection. The model is implemented within ZeaWatch, an AI-powered platform designed to assist farmers and agricultural stakeholders in identifying maize diseases from leaf images. The hybrid architecture leverages CNNs for local feature extraction and Transformers for capturing global contextual relationships, improving classification accuracy across multiple disease classes. The system supports image upload and real-time analysis, providing users with disease identification and recommended interventions. Experimental results demonstrate that the hybrid model outperforms traditional single-architecture approaches in accuracy and robustness. This work contributes to the application of artificial intelligence in precision agriculture, particularly in resource-constrained environments. Keywords: Maize leaf disease, deep learning, CNN, Vision Transformer, agriculture AI, image classification, ZeaWatch.
Joseph Baya Karisa (Mon,) studied this question.