CEBW-YOLO11: A YOLO11-Based Model for Lithium Battery Electrode Sheet Defect Detection

Surface defect detection of lithium battery electrodes is vital for industrial automation but remains challenging due to scale variations and complex backgrounds. To address these challenges, this study proposes CEBW-YOLO11, an enhanced YOLO11n-based framework specifically designed for lithium battery electrode sheet defect detection. The proposed method integrates C3k2-DWR for multi-scale feature extraction, EAC for discriminative feature enhancement, BiFPN-DySample for adaptive feature fusion, and Wise-Inner-IoU for bounding box regression optimization. Experimental results on the electrode defect dataset demonstrate that, compared with the baseline YOLO11n, CEBW-YOLO11 improves Precision, Recall, and mAP@0.5 by 3.8, 4.7, and 3.8 percentage points, respectively. The proposed model achieves an inference speed of 90 FPS at an input resolution of 640 × 640 pixels, showing its potential for real-time electrode defect inspection applications.