Yarn-dyed fabric defect detection based on an improved autoencoder with Fourier convolution

Compared with solid-colored fabrics, the textures in yarn-dyed fabric images are more complex, making the task of defect detection more challenging. To achieve efficient detection, this study proposes an automatic detection framework for dyed fabric defects. The proposed framework consists of a hard...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Textile research journal Jg. 93; H. 5-6; S. 1153 - 1165
Hauptverfasser: Xiang, Jun, Pan, Ruru, Gao, Weidong
Format: Journal Article
Sprache:Englisch
Veröffentlicht: London, England SAGE Publications 01.03.2023
Sage Publications Ltd
Schlagworte:
Ablation
Algorithms
Convolution
Defects
Fabrics
Image acquisition
Image quality
Light sources
Training
convolution autoencoder
yarn-dyed fabric
Fourier convolution
Defect detection
unsupervised learning
ISSN:0040-5175, 1746-7748
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Compared with solid-colored fabrics, the textures in yarn-dyed fabric images are more complex, making the task of defect detection more challenging. To achieve efficient detection, this study proposes an automatic detection framework for dyed fabric defects. The proposed framework consists of a hardware system and a detection algorithm. For efficient and high-quality acquisition of fabric images, an image acquisition assembly equipped with three sets of light sources and a mirror was developed. In addition, a defect detection algorithm based on Fourier convolution and a convolutional autoencoder is proposed. Abandoning the common way of adding noise, this paper proposes to generate image pairs for training using a random masking method in the training phase. In the autoencoder, some traditional convolutional layers are replaced with Fourier convolutional layers. Ablation experiments verify the effectiveness of the mask generation method and Fourier convolution. Compared with other defect detection methods, the proposed method achieves the best performance, which verifies the superiority of the method. The maximum detection speed of the developed system can reach 41 meters per minute, which can meet real-time requirements.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0040-5175
1746-7748
DOI:10.1177/00405175221130519