Deep Learning-Enabled Intensity Level Classification of Corn Rust Disease

Classifying plant diseases quickly and accurately is crucial to reducing their negative effect on agricultural output. In this research, we introduce a deep learning-based strategy for disease severity classification in maize using a multi-layer perceptron (MLP) model. Corn rust, which is brought on by Puccinia fungal infections, is a significant danger to harvest success. The proposed model makes use of a 25,000-image dataset that was painstakingly curated to capture various infection phases and then preprocessed to improve consistency and quality. The MLP framework is used to analyze the complex patterns that characterize fluctuating disease severity. Accuracy, precision, recall, and F1-score are used to assess the model's efficacy across a range of intensities. Understanding the subtleties of classification requires conducting a thorough confusion matrix study. The results of the model are also compared graphically to those of state-of-the-art methods. The results suggest that the model is capable of accurately classifying 96.74% of all intensity levels. Both the confusion matrix and the precision-recall metrics shed light on the strengths of individual classes. The suggested model's competitive position in the landscape of disease categorization algorithms is highlighted by the visual comparison. The ramifications of this study go beyond maize rust and contribute to the development of automated disease identification in agriculture. The combination of deep learning and agricultural knowledge provides a promising path toward better disease management. This study highlights the potential to strengthen food security through data-driven interventions as technology continues to reshape the agricultural landscape.