LDTC-YOLO: A Lightweight Detection Model for Typical Citrus Leaf and Fruit Diseases Under Natural Orchard Conditions

Accurate detection of citrus leaf and fruit diseases is important for precision orchard management. However, natural orchard images often contain small disease symptoms, leaf and fruit overlap, illumination variation, and cluttered backgrounds, making reliable detection challenging. This study proposes LDTC-YOLO (LDTC denotes Lightweight Detection for Typical Citrus Diseases), a lightweight YOLOv8n-based detection model for typical citrus leaf and fruit diseases under natural orchard conditions. To improve detection accuracy and model compactness, LDTC-YOLO integrates an Adaptive Feature Pyramid Network for cross-level feature fusion, Coordinate Attention for disease-region feature enhancement, a Lightweight Shared Convolutional Detection head for reducing parameter redundancy, and Wise-IoU for bounding-box regression optimization. In addition, a self-collected handheld citrus disease dataset, HOCD-4, was constructed using close-range smartphone images captured under natural illumination. The dataset covers leaf and fruit symptoms of four typical citrus diseases: Huanglongbing/citrus greening, black spot, canker, and melanose. On the HOCD-4 test set, averaged across the four disease categories, LDTC-YOLO achieved precision, recall, mAP@0.5, and mAP@0.5:0.95 values of 0.915, 0.843, 0.894, and 0.648, respectively. These results indicate that LDTC-YOLO improves detection performance while maintaining a compact and efficient model profile, providing a potential reference for citrus disease detection under natural orchard conditions; however, its performance on actual mobile or embedded edge devices remains to be validated.