| 严赫,史晓亮,吴超华,曾志,罗威,孙存甫.GMS2-YOLO：融合改进MANet和检测头的金属表面缺陷检测方法[J].电子测量与仪器学报,2026,40(3):176-188 |
| GMS2-YOLO：融合改进MANet和检测头的金属表面缺陷检测方法 |
| GMS2-YOLO: A metal surface defect detection methodintegrating improved MANet and detection head |
| |
| DOI： |
| 中文关键词: 缺陷检测 YOLOv11 特征提取 MSNet SBN检测头 损失函数 |
| 英文关键词:defect detection YOLOv11 feature extraction MSNet SBN Detect loss function |
| 基金项目:国家自然科学基金(52375201)、武汉理工大学产学研科技合作项目(20251hx0411)资助 |
|
|
| Author | Institution |
| Yan He | School of Mechanical and Electrical Engineering, Wuhan University of Technology, Wuhan 430070,
China |
| Shi Xiaoliang | School of Mechanical and Electrical Engineering, Wuhan University of Technology, Wuhan 430070,
China |
| Wu Chaohua | School of Mechanical and Electrical Engineering, Wuhan University of Technology, Wuhan 430070,
China |
| Zeng Zhi | School of Mechanical and Electrical Engineering, Wuhan University of Technology, Wuhan 430070,
China |
| Luo Wei | School of Mechanical and Electrical Engineering, Wuhan University of Technology, Wuhan 430070,
China |
| Sun Cunfu | Wuhan Koide kokan Co., Ltd., Wuhan 430090, China |
|
| 摘要点击次数: 316 |
| 全文下载次数: 250 |
| 中文摘要: |
| 针对工业现场生产中对金属表面缺陷检测存在精度低、误检和漏检的问题以及金属表面缺陷检测面临小目标难以分辨、复杂背景干扰强、噪声影响显著等挑战，尤其在小目标检测与多尺度特征提取方面仍存在不足，针对上述挑战提出GMS2-YOLO模型，以多尺度特征提取、特征融合增强与分离批归一化检测头相结合，提高目标缺陷检测精度。首先，将GhostConv与HGBlock模块结合，将GHGnet网络结构作为主干网络，提升模型对目标缺陷细节特征的提取能力；然后，在颈部结构创新型融合MANet与StarBlock，使用MSNet替换C3k2模块，实现更加多样化和丰富的梯度流信息，增强模型的提取融合能力；其次，重新设计YOLOv11n检测头，采用分离批归一化检测头实现信息在不同层级间的交互与融合，准确识别缺陷目标；最后，使用新的损失函数WIoU，通过提高对中等质量图片的关注度，增强了缺陷检测的精度。实验结果表明，此方法在Self-Dataset数据集上平均精度均值（mAP）mAP@0.5指标达到了80.8%，与YOLOv11n相比提高了3.6%；mAP@0.5：0.95指标达到了53.5%，与YOLOv11n相比提高了8.5%。此外，在NEU-DET数据集中，针对该数据集中的微小缺陷CR，其平均精度（AP）值达到了68.3%，与YOLOv11n相比提高了15.8%。所提模型对缺陷检测精度有较大提升，在解决误检，漏检问题方面有明显优势。此外，与其他主流模型对比，改进模型在没有损失较多推理速度的前提下，提高了检测精度，具有良好的工程应用前景。 |
| 英文摘要: |
| Addressing the issues of low accuracy, false positives, and missed detections in metal surface defect detection in industrial field production, as well as the challenges faced by metal surface defect detection, such as difficulty in distinguishing small targets, strong interference from complex backgrounds, and significant noise impact, especially in the areas of small target detection and multi-scale feature extraction, this study proposes the GMS2-YOLO model to improve the accuracy of target defect detection by combining multi-scale feature extraction, feature fusion enhancement, and a separated batch normalization detection head. Firstly, the GhostConv is combined with the HGBlock module, and the GHGnet network structure is used as the backbone network to improve the model’s ability to extract the detailed features of the target defects. Then, in the innovative fusion of MANet and StarBlock in the neck structure, MSNet is used to replace the C3k2 module to achieve more diversified and rich gradient flow information and enhance the extraction and fusion ability of the model. Secondly, the YOLOv11n detection head is redesigned, and the separated batch normalization detection head is used to realize the interaction and fusion of information at different levels, so as to accurately identify the defect target. Finally, using the new loss function WIoU, the accuracy of defect detection is improved by increasing the attention to medium-quality images. The experimental results show that the mean average precision mAP@0.5 index of this method on the Self-Dataset dataset reaches 80.8%, which is 3.6% higher than that of YOLOv11n. The mAP@0.5：0.95 index reached 53.5%, which was 8.5% higher than that of YOLOv11n. In addition, in the NEU-DET dataset, the AP value of the small defect CR in the dataset reached 68.3%, which was 15.8% higher than that of YOLOv11n. The proposed model greatly improves the accuracy of defect detection, and has obvious advantages in solving the problems of false detection and missed detection. In addition, compared with other mainstream models, the improved model improves the detection accuracy without losing more inference speed, and has good prospects for engineering applications. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
