Detecting of Circular Knitting Fabric Defects Using VGG16 Architecture

Abstract

Although the conventional image processing methods can detect fabric defects, fabric defect detection is an open research problem due to the diversity of defect types. In this paper, the feasibility of VGG16 deep learning architecture for fabric defect detection has been demonstrated. A new fabric defect database is used. The pre-trained model of VGG16 architecture on the new database is built. Thus, the training time of the model is reduced. The experimental results show that the VGG16 model outperforms the traditional Shearlet transform and GLCM methods.

Keywords

• Fabric defect
• Deep learning
• Defect detection

VGG16 Mimarisi Kullanılarak Yuvarlak Örgü Kumaş Hatalarının Tespit Edilmesi

Abstract

Geleneksel görüntü işleme metotları kumaş hatalarını tespit edebilmelerine rağmen kumaş hatası tespiti hata tiplerinin çeşitliliği yüzünden açık bir problemdir. Bu çalışmada VGG16 derin öğrenme mimarisinin kumaş hatası tespiti için uygunluğu gösterilmiştir. Yeni bir kumaş hatası veri tabanı kullanılmıştır. Daha önceden eğitilmiş VGG16 mimarisi veri tabanı üzerinde inşa edilmiştir. Böylece modelin eğitim süresi azaltılmıştır. Deneysel sonuçlar VGG16 modelinin geleneksel Shearlet dönüşümü ve GLCM metotlarından daha iyi olduğunu göstermektedir.

Keywords

• Kumaş hatası
• Derin öğrenme
• Hata tespiti

References

1. Chen M, Yu L, Zhi C, Sun R, Zhu S, Gao Z, et al. Improved faster R-CNN for fabric defect detection based on Gabor filter with Genetic Algorithm optimization. Comput Ind. 2022.134:103551.
2. Hanbay K, Talu MF, Özgüven ÖF. Fabric defect detection systems and methods—A systematic literature review. Optik (Stuttg). 2016 Dec 1;127(24):11960–73.
3. Cuifang Z, Yu C, Jiacheng M. Fabric defect detection algorithm based on PHOG and SVM. Indian J Fibre Text Res. 2020;45:123–6.
4. Zhu D, Pan R, Gao W, Zhang J. Yarn-Dyed fabric defect detection based on autocorrelation function and GLCM. Autex Res J. 2015. 15(3):226–32.
5. Pourkaramdel Z, Fekri-Ershad S, Nanni L. Fabric defect detection based on completed local quartet patterns and majority decision algorithm. Expert Syst Appl. 2022 .198:116827.
6. Zhang B, Tang C. A Method for Defect Detection of Yarn-Dyed Fabric Based on Frequency Domain Filtering and Similarity Measurement. Autex Res J. 2019 Sep 1;19(3):257–62.
7. Vermaak H, Nsengiyumva P, Luwes N. Using the Dual-Tree Complex Wavelet Transform for Improved Fabric Defect Detection. J Sensors. 2016.–8.
8. Hanbay K, Fatih Talu M, Özgüven ÖF, Öztürk D. Real-Time Detection of Knitting Fabric Defects Using Shearlet Transform. Tekst ve Konfeksiyon. 29(1):2019. 3-10.

9. Li F, Li F. Bag of tricks for fabric defect detection based on Cascade R-CNN: https://doi.org/101177/0040517520955229 . 2020.91(5–6):599–612.
10. Liu Z, Huo Z, Li C, Dong Y, Li B. DLSE-Net: A robust weakly supervised network for fabric defect detection. Displays. 2021 Jul 1;68:102008.
11. Ouyang W, Xu B, Hou J, Yuan X. Fabric Defect Detection Using Activation Layer Embedded Convolutional Neural Network. IEEE Access. 2019;7:70130–40.
12. Huang Y, Jing J, Wang Z. Fabric Defect Segmentation Method Based on Deep Learning. IEEE Trans Instrum Meas. 2021;70:1–15.
13. Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition. In: 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings. International Conference on Learning Representations, ICLR; 2014. p. 1–14.