Wheat Crops Disease Prediction Using Artificial Neural Network: In the case of North Wollo in Raya Kobo Weredas

This study explores the development and application of an Artificial Neural Network (ANN) model to predict diseases in wheat crops in the North Wollo region of Ethiopia. With the increasing challenges faced by farmers due to crop diseases, early detection and prediction play a critical role in managing and mitigating losses. The ANN model was designed to identify key diseases affecting wheat, including Wheat Rust, Fusarium Head Blight, Powdery Mildew, Septoria Leaf Spot, Yellow Rust, and Healthy Wheat, using image data collected from the region. The model demonstrated exceptional performance, achieving an overall accuracy of 98%. It also showed strong results across key metrics such as precision, recall, and F1 scores, underscoring its ability to identify disease outbreaks effectively. By predicting these diseases early, the model provides insights that can lead to timely interventions, such as pesticide application or crop rotation, ultimately improving yields and reducing crop losses. Despite the strong performance, the model faced some challenges, particularly in distinguishing healthy wheat from diseased samples and differentiating visually similar diseases like Wheat Rust and Yellow Rust. These limitations suggest the need for further refinement through advanced techniques, such as data augmentation, transfer learning, or exploring deeper neural network architectures. This research demonstrates the potential of AI to improve agricultural sustainability. By providing valuable insights into disease prediction, the ANN model contributes to ongoing global efforts to ensure food security and enhancing agricultural productivity. Future directions could include integrating the model with real-time environmental data from IoT sensors or mobile applications, further improving its accessibility and effectiveness.