skip to main content
Language:
Search Limited to: Search Limited to: Resource type Show Results with: Show Results with: Search type Index

Fabric Defect Detection Using a One-class Classification Based on Depthwise Separable Convolution Autoencoder

Journal of physics. Conference series, 2023-08, Vol.2562 (1), p.12053 [Peer Reviewed Journal]

Published under licence by IOP Publishing Ltd ;Published under licence by IOP Publishing Ltd. This work is published under http://creativecommons.org/licenses/by/3.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. ;ISSN: 1742-6588 ;EISSN: 1742-6596 ;DOI: 10.1088/1742-6596/2562/1/012053

Full text available

Citations Cited by
  • Title:
    Fabric Defect Detection Using a One-class Classification Based on Depthwise Separable Convolution Autoencoder
  • Author: Chen, Chun ; Deng, Xiaoyan ; Yu, Zhuliang ; Wu, Zhengtao
  • Subjects: Anomalies ; Classifiers ; Convolution ; Defects ; Physics
  • Is Part Of: Journal of physics. Conference series, 2023-08, Vol.2562 (1), p.12053
  • Description: Abstract Fabric defect detection is anomaly detection, which is widely studied in the textile industry. Like most anomaly detection tasks, there are some problems hindering detection results, such as class imbalance, defective sample scarcity, and feature selection. This paper proposes a method applying depthwise separable convolution autoencoder on dimensionality reduction and one-class classifier support vector data description (SVDD) to detect fabric defects. A depthwise separable convolution autoencoder can effectively extract sample features with less computation and fewer parameters than the regular convolution, which will be easily used in industrial production. SVDD can only use non-defective samples to train the classifier and solve the difficulty and heavy cost of collecting negative samples (defective samples). In this paper, we will demonstrate the effectiveness of the method on polyester fibers by using accuracy and AUC as evaluation criteria.
  • Publisher: Bristol: IOP Publishing
  • Language: English
  • Identifier: ISSN: 1742-6588
    EISSN: 1742-6596
    DOI: 10.1088/1742-6596/2562/1/012053
  • Source: IOP Publishing Free Content
    IOPscience (Open Access)
    GFMER Free Medical Journals
    ProQuest Central
Back to results list
INSPIRE LIBRARY - TON DUC THANG UNIVERSITY (84-028) 37 755 057 Feedback
19 Nguyen Huu Tho St. Dist.7, HCM thuvien@tdtu.edu.vn

Copyright © 2017 INSPIRE LIBRARY - TON DUC THANG UNIVERSITY

Searching Remote Databases, Please Wait