Manuscript received March 20, 2025; revised June 5, 2025; accepted July 10, 2025; published August 7, 2025.

Abstract—Corn is one of the major staple crops globally, particularly in the Philippines, where agricultural productivity is increasingly threatened by diseases such as Northern Corn Leaf Blight, Gray Leaf Spot, and Corn Rust. These challenges are further exacerbated by tropical cyclones and fluctuating environmental conditions, leading to substantial losses in yield and crop quality. This study presents CLDS-YOLO (Corn Leaf Disease Detection and Severity Evaluation Using YOLOv9), a novel system that leverages YOLOv9 instance segmentation for accurate disease detection and integrates fuzzy logic for severity assessment, based on Relative Leaf Area (RLA) and the count of diseased regions. The YOLOv9e-seg model demonstrated strong performance across classification tasks, achieving an overall accuracy of 80%, with recall values of 85% for diseased regions, 80% for healthy leaf areas, and 75% for background, based on the normalized confusion matrix. Precision levels were similarly high, particularly for leaf detection, while the model maintained a balanced trade-off in identifying diseased and background classes. These improvements address previous segmentation challenges and confirm the model’s robustness in multi-class classification. Furthermore, the severity analysis effectively categorized disease levels, supporting timely and informed crop management decisions. The CLDS-YOLO system demonstrates significant potential for real-time disease detection and severity evaluation, laying the groundwork for an indoor planting framework that ensures continuous health monitoring and protection from adverse weather conditions.

Keywords—corn leaf disease, disease detection, YOLOv9, instance segmentation, computer vision, relative leaf area, fuzzy logic, severity evaluation, agriculture technology

Cite: Elmo Ranolo, Ken Gorro, Lawrence Roble, Christia Mae Camay, Rue Nicole Santillan, Anthony Ilano, Deofel Balijon, and Daniel Ariaso Sr., "CLDS-YOLO: Corn Leaf Disease Detection and Severity Evaluation Using YOLOv9," Journal of Image and Graphics, Vol. 13, No. 4, pp. 437-451, 2025.

Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC-BY-4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.