Fig. 15
From: An enhanced deep learning-based framework for diagnosing apple leaf diseases
Qualitative analysis of E-YOLOv8 on challenging cases. Row 1: under low-light conditions, the model successfully detects lesions despite reduced contrast. Row 2: With overlapping leaves, E-YOLOv8 localizes disease regions even under partial occlusion. Row 3: for small lesions, the refined FPN and CBAM modules enable accurate detection of subtle symptoms. These examples demonstrate the robustness of the proposed method in real-world agricultural scenarios.
