This is the authors' abstract. We don't add key points for this paper.
In production, defect detection is crucial for ensuring product quality and consumer satisfaction. To address the issues of surface defect detection in liquor bottle caps and the large number of algorithm parameters, this study improves YOLO11n and proposes a more lightweight and higher-precision HP-YOLO11n algorithm. Firstly, We employ the improved HGNetv2 backbone network as our model backbone, which makes the model more lightweight while ensuring the accuracy of model detection. Secondly, added a P2 detection layer to YOLO11n, incorporating high-resolution feature maps and rich detailed information to enhance the model's overall recognition performance. Finally, we remove the P5 layer used for detecting large targets, which reduces the number of parameters and computational load while maintaining accuracy. The experimental results show that the HP-YOLO11n algorithm achieves a mean average precision mAP@0.5 of 87.06%, which is 1.52 percentage points higher than the original YOLOv11n algorithm, while reducing the number of parameters by 44.57%, making it more accurate and lightweight.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: