arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2402.02797cs.CVcs.LG
收藏

联合注意力引导特征融合网络用于表面缺陷显著性检测

Joint Attention-Guided Feature Fusion Network for Saliency Detection of Surface Defects

Zhengzhou University(郑州大学);University of Science and Technology of China(中国科学技术大学);Tianjin University(天津大学);Beihang University(北京航空航天大学)

Xiaoheng Jiang, Feng Yan, Yang Lu, Ke Wang, Shuai Guo, Tianzhu Zhang, Yanwei Pang, Jianwei Niu, Mingliang Xu

AI总结:

提出JAFFNet网络,通过联合注意力引导特征融合和密集感受野模块,解决表面缺陷检测中的尺度变化、复杂背景和低对比度问题,在多个数据集上取得优异性能并实现66 FPS实时检测。

摘要:

表面缺陷检测在工业生产制造过程中扮演着重要角色。尽管基于卷积神经网络(CNN)的缺陷检测方法已经取得了巨大进步,但它们仍然面临许多挑战,如缺陷尺度变化、复杂背景、低对比度等。为了解决这些问题,我们提出了一种基于编码器-解码器网络的联合注意力引导特征融合网络(JAFFNet),用于表面缺陷的显著性检测。JAFFNet主要将联合注意力引导特征融合(JAFF)模块集成到解码阶段,以自适应地融合低级和高级特征。JAFF模块在特征融合过程中学习强调缺陷特征并抑制背景噪声,这有利于检测低对比度缺陷。此外,JAFFNet在编码器之后引入了一个密集感受野(DRF)模块,以捕获具有丰富上下文信息的特征,这有助于检测不同尺度的缺陷。JAFF模块主要利用由高级语义特征提供的学习到的联合通道-空间注意力图来引导特征融合。该注意力图使模型更加关注缺陷特征。DRF模块利用一系列多感受野(MRF)单元,每个单元将前面所有MRF特征图和原始输入作为输入。获得的DRF特征捕获了具有大范围感受野的丰富上下文信息。在SD-saliency-900、Magnetic tile和DAGM 2007数据集上进行的大量实验表明,与其他最先进的方法相比,我们的方法取得了有前景的性能。同时,我们的方法达到了66 FPS的实时缺陷检测速度。

英文摘要:

Surface defect inspection plays an important role in the process of industrial manufacture and production. Though Convolutional Neural Network (CNN) based defect inspection methods have made huge leaps, they still confront a lot of challenges such as defect scale variation, complex background, low contrast, and so on. To address these issues, we propose a joint attention-guided feature fusion network (JAFFNet) for saliency detection of surface defects based on the encoder-decoder network. JAFFNet mainly incorporates a joint attention-guided feature fusion (JAFF) module into decoding stages to adaptively fuse low-level and high-level features. The JAFF module learns to emphasize defect features and suppress background noise during feature fusion, which is beneficial for detecting low-contrast defects. In addition, JAFFNet introduces a dense receptive field (DRF) module following the encoder to capture features with rich context information, which helps detect defects of different scales. The JAFF module mainly utilizes a learned joint channel-spatial attention map provided by high-level semantic features to guide feature fusion. The attention map makes the model pay more attention to defect features. The DRF module utilizes a sequence of multi-receptive-field (MRF) units with each taking as inputs all the preceding MRF feature maps and the original input. The obtained DRF features capture rich context information with a large range of receptive fields. Extensive experiments conducted on SD-saliency-900, Magnetic tile, and DAGM 2007 indicate that our method achieves promising performance in comparison with other state-of-the-art methods. Meanwhile, our method reaches a real-time defect detection speed of 66 FPS.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑