Maize Leaf Disease Detection Using a Transfer Learning Approach

Maize, as a staple food crop, plays a crucial role in global food security. However, its production is significantly threatened by various diseases affecting maize leaves. In order to reduce yield expenses, early identification and treatment of certain diseases become crucial. In this work, we suggest a transfer learning model for the diagnosis of maize crop leaf disease. In this paper used pre-trained models such as MobileNet and MobileNetV2, fine-tuned on a dataset comprising images of maize leaves with different diseases. The final results demonstrate the effectiveness of transfer learning, with the suggested model achieving an accuracy of 96.00% on the PlantVillage dataset, outperforming existing methods. This approach shows great potential for practical application in agriculture, enabling early and accurate detection of maize leaf diseases for timely manner and enhancing the crop yield.