Investigating Shift Equivalence of Convolutional Neural Networks in Industrial Defect Segmentation

CASIA OpenIR > 中国科学院工业视觉智能装备工程实验室 > 精密感知与控制

	Investigating Shift Equivalence of Convolutional Neural Networks in Industrial Defect Segmentation
	Qu Z(屈震)1 ; Tao X(陶显)1 ; Shen F(沈飞)1 ; Zhang ZT(张正涛)1 ; Li T(李涛)2
发表期刊	IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
	2023-10
页码	1-17
摘要	In industrial defect segmentation tasks, while pixel accuracy and Intersection over Union (IoU) are commonly employed metrics to assess segmentation performance, the output consistency (also referred to as equivalence) of the model is often overlooked. Even a small shift in the input image can yield significant fluctuations in the segmentation results. Existing methodologies primarily focus on data augmentation or antialias ing to enhance the network’s robustness against translational transformations, but their shift equivalence performs poorly on the test set or is susceptible to nonlinear activation functions. In addition, the variations in boundaries resulting from the trans lation of input images are consistently disregarded, thus imposing further limitations on the shift equivalence. In response to this particular challenge, a novel pair of downsampling/upsampling layers called component attention polyphase sampling (CAPS) is proposed as a replacement for the conventional sampling layers in CNNs. To mitigate the effect of image boundary variations on the equivalence, an adaptive windowing module is designed in CAPS to adaptively filter out the border pixels of the image. Furthermore, a component attention module is proposed to fuse all downsampled features to improve the segmentation performance. The experimental results on the micro surface defect (MSD) dataset and four real-world industrial defect datasets demonstrate that the proposed method exhibits higher equivalence and segmentation performance compared to other state-of-the-art methods. Our code will be available at https://github.com/xiaozhen228/CAPS.
收录类别	SCI
语种	英语
七大方向——子方向分类	人工智能+制造
国重实验室规划方向分类	先进智能应用与转化
是否有论文关联数据集需要存交	否
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/57185
专题	中国科学院工业视觉智能装备工程实验室_精密感知与控制
通讯作者	Qu Z(屈震); Tao X(陶显)
作者单位	1.Institute of Automation, Chinese Academy of Sciences 2.Institute of Automation, Gansu Academy of Sciences
第一作者单位	中国科学院自动化研究所
通讯作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Qu Z,Tao X,Shen F,et al. Investigating Shift Equivalence of Convolutional Neural Networks in Industrial Defect Segmentation[J]. IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,2023:1-17.
APA	Qu Z,Tao X,Shen F,Zhang ZT,&Li T.(2023).Investigating Shift Equivalence of Convolutional Neural Networks in Industrial Defect Segmentation.IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,1-17.
MLA	Qu Z,et al."Investigating Shift Equivalence of Convolutional Neural Networks in Industrial Defect Segmentation".IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2023):1-17.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Investigating_Shift_（2869KB）	期刊论文	作者接受稿	开放获取	CC BY-NC-SA	浏览下载