Efficient and accurate identification of maize rust disease using deep learning model

Common corn rust and southern corn rust, two typical maize diseases during growth stages, require accurate differentiation to understand their occurrence patterns and pathogenic risks. To address this, a specialized Maize-Rust model integrating a SimAM module in the YOLOv8s backbone and a BiFPN for scale fusion, along with a DWConv for streamlined detection, was developed. The model achieved an accuracy of 94.6%, average accuracy of 91.6%, recall rate of 85.4%, and F1 value of 0.823, outperforming Faster-RCNN and SSD models by 16.35% and 12.49% in classification accuracy, respectively, and detecting a single rust image at 16.18 frames per second. Deployed on mobile phones, the model enables real-time data collection and analysis, supporting effective detection and management of large-scale outbreaks of rust in the field.