For the sake of recognizing and classifying textile defects, deep learning-based methods have been proposed and have achieved remarkable success in single-label textile images. However, detecting multilabel defects in a textile image remains challenging due to the coexistence of multiple defects and small-size defects. To address these challenges, a multilevel, multi-attentional deep learning network was proposed and built to: (a) increase the feature representation ability to detect small-size defects; and (b) generate discriminative representation that maximizes the capability of attending the defect status, which leverages higher-resolution feature maps for multiple defects. Moreover, a multilabel object detection dataset in textile defect images was built to verify the performance of the proposed model. The results demonstrate that the network extracts more distinctive features and has better performance than the state-of-the-art approaches on the real-world industrial dataset.
WeiBHaoKGaoL, et al.
Bioinspired visual-integrated model for multilabel classification of textile defect images. IEEE Trans Cogn Dev Syst2021;
13: 503–513.
3.
WeiBHeHHaoK, et al.
Visual interaction networks: a novel bio-inspired computational model for image classification. Neural Networks2020;
130: 100–110.
4.
WeiBHaoKGaoL, et al.
Detecting textile micro-defects: a novel and efficient method based on visual gain mechanism. Inform Sci2020;
541: 60–74.
5.
BuHGHuangXBWangJ, et al.
Detection of fabric defects by auto-regressive spectral analysis and support vector data description. Text Res J2010;
80: 579–589.
6.
PanRGaoWLiuJ, et al. Automatic detection of structure parameters of yarn-dyed fabric. Text Res J2010;
80: 1819–1832.
7.
ZhengD.A new method for the detection and classification of weave pattern repeat. Text Res J2014;
84: 1586–1599.
8.
ZhouJWangJ.Fabric defect detection using adaptive dictionaries. Text Res J2013;
83: 1846–1859.
9.
JunXWangJZhouJ, et al.
Fabric defect detection based on a deep convolutional neural network using a two-stage strategy. Text Res J2021;
91: 130–142.
10.
HuGHuangJWangQ, et al.
Unsupervised fabric defect detection based on a deep convolutional generative adversarial network. Text Res J2020;
90: 247–270.
11.
WeiBHaoKTangXS, et al.
Fabric defect detection based on faster RCNN. International Conference on Artifical Intelligence on Textile and Apparel (AIFT), 3–6, July, 2018, Hong Kong, China, pp. 45–51.
12.
ChangXGuCLiangJ, et al.
Fabric defect detection based on pattern template correction. Math Probl Eng2018;
2018: 1–17.
13.
HuangpengQZhangHZengX, et al.
Automatic visual defect detection using texture prior and low-rank representation. IEEE Access2018;
6: 37965–37976.
14.
SongMZhouXCaiY, et al. Algorithm for cheese yarn defects detecting based on multi-directional matching filter. International Conference on Image, Vision and Computing (ICIVC), 27–29, June, 2018, Chongqing, China, pp. 696–700. IEEE.
15.
GuanSShiH.Fabric defect detection based on the saliency map construction of target-driven feature. J Text Inst2018;
109: 1133–1142.
HanTHaoKDingY, et al.
A sparse autoencoder compressed sensing method for acquiring the pressure array information of clothing. Neurocomputing2018;
275: 1500–1510.
18.
WeiBHaoKTangX, et al.
A new method using the convolutional neural network with compressive sensing for fabric defect classification based on small sample sizes. Text Res J2019;
89: 3539–3555.
19.
MeiSYangHYinZ.An unsupervised-learning-based approach for automated defect inspection on textured surfaces. IEEE Trans Instrum Meas2018;
67: 1266–1277.
20.
ToyouraMIgarashiTMaoX.Generating jacquard fabric pattern with visual impressions. IEEE Trans Ind Informatics2018; 15(8): 4536–4544.
21.
LvSShiSWangH, et al.
Semi-supervised multi-label feature selection with adaptive structure learning and manifold learning. Knowledge-Based Syst2021;
214: 106757.
22.
ZhaoDGaoQLuY, et al.
Consistency and diversity neural network multi-view multi-label learning. Knowledge-Based Syst2021;
218: 106841.
23.
FanYLiuJWengW, et al.
Multi-label feature selection with constraint regression and adaptive spectral graph. Knowledge-Based Syst2021;
212: 106621.
24.
YaoCSunPZhiR, et al.
Learning coexistence discriminative features for multi-class object detection. IEEE Access2018;
6: 37676–37684.
25.
GuanSGaoZ.Fabric defect image segmentation based on the visual attention mechanism of the wavelet domain. Text Res J2014;
84: 1018–1033.
26.
NogueiraKPenattiOABdos SantosJA.Towards better exploiting convolutional neural networks for remote sensing scene classification. Pattern Recognit2017;
61: 539–556.
AdelsonEHBurtPJAndersonCH, et al. Pyramid methods in image processing. RCA Eng1984;
29: 33–41.
29.
LiuWAnguelovDErhanD, et al.
SSD: single shot multiBox detector.Proc Eur Conf Comput Vis2016; 9905: 21–37.
30.
LinTDollPGirshickR, et al.
Feature pyramid networks for object detection. 2017 IEEE Conf Comput Vis Pattern Recognit2017; 1(2): 2117–2125.
31.
SindagiVPatelV.Multi-level bottom-top and top-bottom feature fusion for crowd counting. Proc IEEE Int Conf Comput Vis2019; 27: 1002–1012.
32.
ZhangPWangDLuH, et al. Amulet : Aggregating Multi-level Convolutional Features. Proc IEEE Int Conf Comput Vis2017; 31: 202–211.
33.
RensinkRA.The dynamic representation of scenes. Vis Cognit2000;
7: 17–42.
34.
VaswaniAShazeerNParmarN, et al.
Attention is all you need. Adv Neural Inf Process Syst2017; 30: 5999–6009.
35.
LiangHJSanketNJFermullerC, et al.
SalientDSO: bringing attention to direct sparse odometry. IEEE Trans Autom Sci Eng2019;
16: 1619–1626.
36.
WeiBHaoKGaoL, et al. Bio-inspired visual integrated model for multi-label classification of textile defect images. IEEE Trans Cogn Dev Syst2020; 13(3): 503–513.
37.
WangWShenJ.Deep visual attention prediction. IEEE Trans Image Process2018;
27: 2368–2378.
38.
ZhuYZhaoCGuoH, et al.
Attention CoupleNet: fully convolutional attention coupling network for object detection. IEEE Trans Image Process2019;
28: 113–126.
39.
XuKCourvilleAZemelRS, et al. Show, Attend and Tell : Neural Image Caption Generation with Visual Attention. In: International Conference on Machine Learning. 6–11, July, 2015, Lille, France, pp. 2048–2057.
40.
KrizhevskyAHintonGE.ImageNet classification with deep convolutional neural networks. Proc Adv Neural Inf Process Syst2012; 60: 1–9.
41.
HeKZhangXRenS, et al. Deep residual learning for image recognition. Proc IEEE Conf Comput Vis Pattern Recognit2016; 2016: 770–778.
42.
ZhangMZhouZ.A review on multi-label learning algorithms. IEEE Trans Knowl Data Eng2013;
26: 1819–1837.
43.
SimonyanKZissermanA.Very deep convolutional networks for large-scale image recognition. Proc Int Conf Learn Represent2018;
75: 398–406.
44.
Cornell University. RedmonJFarhadiA.YOLOv3: An Incremental Improvement. (2018). http://arxiv.org/abs/1804.02767 (accessed 14 January 2022).
45.
GirshickRDonahueJDarrellT, et al. Rich feature hierarchies for accurate object detection and semantic segmentation. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit2014; 24: 580–587.
46.
RenSHeKGirshickR, et al. Faster R-CNN : Towards Real-Time Object Detection with Region Proposal Networks. In: Proc. 28th Int. Conf. Neural Inf. Process. Syst. 7–12, December, 2015, Montreal, QC, Canada, pp. 1–14.
