| 赵云亮,唐东林,何媛媛,丁超,杨洲.基于CNN融合PGW-Attention的金属表面缺陷识别方法[J].电子测量与仪器学报,2024,38(8):46-55 |
| 基于CNN融合PGW-Attention的金属表面缺陷识别方法 |
| Metal surface defect recognition method based on CNN with PGW-Attention |
| |
| DOI： |
| 中文关键词: 金属表面缺陷 CNN Transformer 深度卷积 PGW Attention |
| 英文关键词:metal surface defects CNN Transformer deep convolution PGW-Attention |
| 基金项目:国家市场监督管理总局科技计划项目(2022MK115)、四川省市场监督管道局科技计划项目（CSCJZ2022007）资助 |
|
|
| Author | Institution |
| Zhao Yunliang | School of Electrical and Mechanical Engineering, Southwest Petroleum University, Chengdu 610500,China |
| Tang Donglin | School of Electrical and Mechanical Engineering, Southwest Petroleum University, Chengdu 610500,China |
| He Yuanyuan | Sichuan Special Equipment Inspection Institute, Chengdu 610061,China |
| Ding Chao | School of Intelligent Manufacturing,
Chengdu Technological University, Chengdu 611730,China |
| Yang Zhou | School of Electrical and Mechanical Engineering, Southwest Petroleum University, Chengdu 610500,China |
|
| 摘要点击次数: 37 |
| 全文下载次数: 104 |
| 中文摘要: |
| 针对分散和细小的金属表面缺陷检测方法,卷积神经网络(CNN)缺乏全局特征捕捉能力,在识别氧化颗粒、裂纹和划痕等缺陷时易发生漏检和特征丢失,Transformer能够捕捉图像全局信息,但全局计算导致较高的计算成本。为实现高效且精准的金属表面缺陷识别,将CNN的局部特征提取能力与Transformer的全局建模能力有效融合,提出了一种基于深度可分离卷积(DW-Conv)融合池化网格窗口注意力机制(PGW-Attention)的金属表面缺陷识别网络架构(DPG-Transformer)。在自建金属缺陷数据集(ST-DET)和公开金属缺陷数据集(NEU-CLS)上对该方法进行了实验验证,DPG-Transformer的缺陷识别准确率分别为99.3%和99.6%,在准确率、计算量和浮点计算量等指标上优于多种经典网络。此外,在可视化实验中,DPG-Transformer显示出比CNN模型更全面地腐蚀和氧化皮的缺陷特征提取能力,并能比Transformer模型更加精准地关注到细长裂纹和划痕的全局缺陷特征。实验结果表明,该方法可以降低Transformer模型的计算量和复杂度,同时能够更全面、精准地提取到金属表面缺陷特征,是一种更切合实际应用的金属表面缺陷检测方法。 |
| 英文摘要: |
| To address the challenges in detecting dispersed and fine defects on metal surfaces, convolutional neural network (CNN) often fall short due to their limited ability to capture global features, leading to missed detections and loss of detail in identifying defects such as oxidation particles, cracks, and scratches. Although Transformers can capture comprehensive global information, the extensive computation required can be costly. In pursuit of an efficient and accurate method for metal surface defect detection, this study introduces a novel network architecture, the DPG-Transformer, which synergistically combines the local feature extraction capabilities of CNNs with the global modeling strengths of Transformer. This integration is facilitated through the use of depthwise separable convolutions (DW-Conv) and pooling grid window attention mechanisms (PGW-Attention). The effectiveness of the DPG-Transformer was validated on both a proprietary metal defect dataset (ST-DET) and a public dataset (NEU-CLS), achieving defect detection accuracies of 99.3% and 99.6%, respectively, and outperforming several classic networks in terms of accuracy, computational efficiency, and floating-point operations. Additionally, visualization experiments demonstrated that the DPG-Transformer more comprehensively extracts defect features associated with corrosion and scaling compared to CNN models, and more precisely focuses on the global features of elongated cracks and scratches than Transformer models. The results indicate that the DPG-Transformer not only reduces computational load and complexity but also enhances the comprehensive and precise detection of metal surface defects, making it a highly suitable approach for practical applications in metal surface defect detection. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
