Algorithm evaluation demonstrates improved detection accuracy and lower computational load for electric vehicle battery air-tightness, suggesting an efficient framework for automated manufacturing.
With the rapid development of the new energy vehicle industry, battery air‐tightness inspection faces higher requirements. Traditional methods suffer from low efficiency and inaccurate detection. This paper proposes a visual inspection method based on improved YOLOv11, achieving reduced model complexity and enhanced inspection accuracy by introducing a lightweight MobileNetV4 network as the backbone feature extractor and optimizing its convolutional layers with the Ghost Convolution (GhostConv) module. Experimental results on a self‐built battery pack leakage dataset show that the improved model achieves 98.7% in mAP@0.5 and 79.2% in mAP@0.5:0.95; compared with the base model YOLOv11n, its mAP@0.5 and mAP@0.5:0.95 improve by 1.2 percentage points and 1.6 percentage points respectively, with a 13.6% reduction in FLOPs. This algorithm demonstrates higher accuracy and lighter weight than other YOLO series methods, providing an innovative solution for efficient and accurate intelligent inspection of battery air‐tightness in new energy vehicles.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: