Abstract

This paper proposes an approach for automated defect detection in homogeneous textiles using texture analysis. The texture features are extracted by the wavelet packet frame decomposition followed by the Karhunen–Loève transform. The texture feature vector for each pixel is used as an input to a Gaussian mixture model that determines whether or not each pixel is defective. The parameters of the Gaussian mixture model are estimated with nondefective textile images in supervised defect detection. An approach for unsupervised defect detection is also presented that can identify the heterogeneous subblocks on the basis of the Kullback–Leibler divergence between two Gaussian mixtures. The proposed method was evaluated on 25 different homogeneous textile image pairs, one of each pair with a defect and the other with no defect, and was compared with existing methods using texture analysis. The experimental results yielded visually good segmentation and an excellent detection rate with a low false alarm rate for both supervised and unsupervised defect detection. This confirms the validity of the proposed approach for automated defect detection and localization.

© 2006 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. K. Schicktanz, "Automatic fault detection possibilities on nonwoven fabrics," Melliand Textilber. 74, 294-295 (1993).
  2. H. S. Don, K. S. Fu, C. R. Liu, and W. C. Lin, "Metal surface inspection using image processing techniques," IEEE Trans. Syst. Man Cybern. 3, 139-146 (1984).
  3. F. S. Cohen, Z. Fan, and S. Attali, "Automated inspection of textile fabrics using textural models," IEEE Trans. Pattern Anal. Mach. Intell. 13, 803-808 (1991).
    [CrossRef]
  4. C. H. Chan and G. K. H. Pang, "Fabric defect detection by Fourier analysis," IEEE Trans. Ind. Appl. 36, 1267-1276 (2000).
    [CrossRef]
  5. E. J. Wood, "Applying Fourier and associated transforms to pattern characterization in textile," Text. Res. J. 60, 212-220 (1990).
    [CrossRef]
  6. A. Bodnarova, M. Bennamoun, and S. Latham, "Optimal Gabor filters for textile flaw detection," Pattern Recogn. 35, 2973-2991 (2002).
    [CrossRef]
  7. W. J. Jasper and H. Potlapalli, "Image analysis of mispicks in woven fabric," Text. Res. J. 65, 683-692 (1995).
    [CrossRef]
  8. X. Yang, G. K. H. Pang, and N. Yung, "Discriminative fabric defect detection using adaptive wavelet," Opt. Eng. (Bellingham) 41, 3116-3226 (2002).
    [CrossRef]
  9. T. Chang and C. C. J. Kuo, "Texture analysis and classification with tree-structured wavelet transform," IEEE Trans. Image Process. 2, 429-441 (1993).
    [CrossRef] [PubMed]
  10. A. Laine and J. Fan, "Texture classification by wavelet packet signatures," IEEE Trans. Pattern Anal. Mach. Intell. 15, 1186-1191 (1993).
    [CrossRef]
  11. C. S. Lu, P. C. Chung, and C. F. Chen, "Unsupervised texture segmentation via wavelet transform," Pattern Recogn. 30, 729-742 (1997).
    [CrossRef]
  12. M. Acharyya and M. K. Kundu, "An adaptive approach to unsupervised texture segmentation using m-band wavelet transform," Signal Process. 81, 1337-1356 (2001).
    [CrossRef]
  13. M. Unser, "Texture classification and segmentation using wavelet frames," IEEE Trans. Image Process. 4, 1549-1560 (1995).
    [CrossRef] [PubMed]
  14. R. Cossu, I. H. Jermyn, and J. Zerubia, "Texture analysis using probabilistic models of the unimodal and multimodal statistics of adaptive wavelet packet coefficients," in Proceedings of IEEE 2004 International Conference on Acoustics, Speech, and Signal Processing (IEEE, 2004), pp. 597-600.
  15. G. Van de Wouwer, P. Scheunders, and D. Van Dyck, "Statistical texture characterization from discrete wavelet representations," IEEE Trans. Image Process. 8, 592-598 (1999).
    [CrossRef]
  16. S. C. Kim and T. J. Kang, "Texture classification and segmentation using wavelet packet frame and Gaussian mixture model," Pattern Recogn. (to be published).
  17. K. K. Yiu, M. W. Mak, and C. K. Li, "Gaussian mixture models and probabilistic decision-based neural networks for pattern classification: a comparative study," Neural Comput. Appl. 8, 235-245 (1999).
    [CrossRef]
  18. A. P. Dempster, N. M. Laird, and D. B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. R. Stat. Soc. Ser. B. Methodol. 39, 1-38 (1977).
  19. C. L. Tseng, Y. H. Chen, Y. Y. Xu, H. T. Pao, and H. C. Fu, "A self-growing probabilistic decision-based neural network with automatic data clustering," Ind. Math. 61, 21-38 (2004).
  20. J. Escofet, R. Navarro, M. S. Millan, and J. Pladellorens, "Detection of local defects in textiles webs using Gabor filters," Opt. Eng. (Bellingham) 37, 2297-2307 (1998).
    [CrossRef]
  21. A. Kumar and G. K. H. Pang, "Defect detection in textured materials using Gabor filters," IEEE Trans. Ind. Appl. 38, 425-440 (2002).
    [CrossRef]
  22. A. Kumar and G. K. H. Pang, "Defect detection in textured materials using optimized filters," IEEE Trans. Syst. Man Cybern. Part B: Cybern. 32(5), 553-570 (2002).
    [CrossRef]
  23. C. Anagnostopoulos, D. Vergados, E. Kayafas, V. Loumos, and G. Stassinopoulos, "A computer vision approach for textile quality control," J. Vis. Comput. Animat. 12, 31-44 (2001).
    [CrossRef]
  24. A. Abouelela, H. M. Abbas, H. Eldeeb, A. A. Wahdan, and S. M. Nassar, "Automated vision system for localizing structural defects in textile fabrics," Pattern Recogn. Lett. 26, 1435-1443 (2005).
    [CrossRef]
  25. A. K. Jain and F. Furrokhnia, "Unsupervised texture segmentation using Gabor filters," Pattern Recogn. 23, 1167-1186 (1991).
    [CrossRef]
  26. B. Xu, "Identifying fabric structure with fast Fourier transform techniques," Text. Res. J. 66, 496-506 (1996).
    [CrossRef]
  27. J. Goldberger, S. Gordon, and H. Greenspan, "An efficient image similarity measure based on approximations of KL divergence between two Gaussian mixtures," in Proceedings of IEEE Ninth International Conference on Computer Vision (IEEE, 2003), pp. 487-493.
    [CrossRef]
  28. A. Bodnarova, M. Bennamoun, and K. K. Kubik, "Suitability analysis of techniques for flaw detection in textiles using texture analysis," Pattern Anal. Appl. 3, 254-266 (2000).
    [CrossRef]

2005 (1)

A. Abouelela, H. M. Abbas, H. Eldeeb, A. A. Wahdan, and S. M. Nassar, "Automated vision system for localizing structural defects in textile fabrics," Pattern Recogn. Lett. 26, 1435-1443 (2005).
[CrossRef]

2004 (1)

C. L. Tseng, Y. H. Chen, Y. Y. Xu, H. T. Pao, and H. C. Fu, "A self-growing probabilistic decision-based neural network with automatic data clustering," Ind. Math. 61, 21-38 (2004).

2002 (4)

A. Bodnarova, M. Bennamoun, and S. Latham, "Optimal Gabor filters for textile flaw detection," Pattern Recogn. 35, 2973-2991 (2002).
[CrossRef]

X. Yang, G. K. H. Pang, and N. Yung, "Discriminative fabric defect detection using adaptive wavelet," Opt. Eng. (Bellingham) 41, 3116-3226 (2002).
[CrossRef]

A. Kumar and G. K. H. Pang, "Defect detection in textured materials using Gabor filters," IEEE Trans. Ind. Appl. 38, 425-440 (2002).
[CrossRef]

A. Kumar and G. K. H. Pang, "Defect detection in textured materials using optimized filters," IEEE Trans. Syst. Man Cybern. Part B: Cybern. 32(5), 553-570 (2002).
[CrossRef]

2001 (2)

C. Anagnostopoulos, D. Vergados, E. Kayafas, V. Loumos, and G. Stassinopoulos, "A computer vision approach for textile quality control," J. Vis. Comput. Animat. 12, 31-44 (2001).
[CrossRef]

M. Acharyya and M. K. Kundu, "An adaptive approach to unsupervised texture segmentation using m-band wavelet transform," Signal Process. 81, 1337-1356 (2001).
[CrossRef]

2000 (2)

C. H. Chan and G. K. H. Pang, "Fabric defect detection by Fourier analysis," IEEE Trans. Ind. Appl. 36, 1267-1276 (2000).
[CrossRef]

A. Bodnarova, M. Bennamoun, and K. K. Kubik, "Suitability analysis of techniques for flaw detection in textiles using texture analysis," Pattern Anal. Appl. 3, 254-266 (2000).
[CrossRef]

1999 (2)

G. Van de Wouwer, P. Scheunders, and D. Van Dyck, "Statistical texture characterization from discrete wavelet representations," IEEE Trans. Image Process. 8, 592-598 (1999).
[CrossRef]

K. K. Yiu, M. W. Mak, and C. K. Li, "Gaussian mixture models and probabilistic decision-based neural networks for pattern classification: a comparative study," Neural Comput. Appl. 8, 235-245 (1999).
[CrossRef]

1998 (1)

J. Escofet, R. Navarro, M. S. Millan, and J. Pladellorens, "Detection of local defects in textiles webs using Gabor filters," Opt. Eng. (Bellingham) 37, 2297-2307 (1998).
[CrossRef]

1997 (1)

C. S. Lu, P. C. Chung, and C. F. Chen, "Unsupervised texture segmentation via wavelet transform," Pattern Recogn. 30, 729-742 (1997).
[CrossRef]

1996 (1)

B. Xu, "Identifying fabric structure with fast Fourier transform techniques," Text. Res. J. 66, 496-506 (1996).
[CrossRef]

1995 (2)

M. Unser, "Texture classification and segmentation using wavelet frames," IEEE Trans. Image Process. 4, 1549-1560 (1995).
[CrossRef] [PubMed]

W. J. Jasper and H. Potlapalli, "Image analysis of mispicks in woven fabric," Text. Res. J. 65, 683-692 (1995).
[CrossRef]

1993 (3)

T. Chang and C. C. J. Kuo, "Texture analysis and classification with tree-structured wavelet transform," IEEE Trans. Image Process. 2, 429-441 (1993).
[CrossRef] [PubMed]

A. Laine and J. Fan, "Texture classification by wavelet packet signatures," IEEE Trans. Pattern Anal. Mach. Intell. 15, 1186-1191 (1993).
[CrossRef]

K. Schicktanz, "Automatic fault detection possibilities on nonwoven fabrics," Melliand Textilber. 74, 294-295 (1993).

1991 (2)

F. S. Cohen, Z. Fan, and S. Attali, "Automated inspection of textile fabrics using textural models," IEEE Trans. Pattern Anal. Mach. Intell. 13, 803-808 (1991).
[CrossRef]

A. K. Jain and F. Furrokhnia, "Unsupervised texture segmentation using Gabor filters," Pattern Recogn. 23, 1167-1186 (1991).
[CrossRef]

1990 (1)

E. J. Wood, "Applying Fourier and associated transforms to pattern characterization in textile," Text. Res. J. 60, 212-220 (1990).
[CrossRef]

1984 (1)

H. S. Don, K. S. Fu, C. R. Liu, and W. C. Lin, "Metal surface inspection using image processing techniques," IEEE Trans. Syst. Man Cybern. 3, 139-146 (1984).

1977 (1)

A. P. Dempster, N. M. Laird, and D. B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. R. Stat. Soc. Ser. B. Methodol. 39, 1-38 (1977).

Abbas, H. M.

A. Abouelela, H. M. Abbas, H. Eldeeb, A. A. Wahdan, and S. M. Nassar, "Automated vision system for localizing structural defects in textile fabrics," Pattern Recogn. Lett. 26, 1435-1443 (2005).
[CrossRef]

Abouelela, A.

A. Abouelela, H. M. Abbas, H. Eldeeb, A. A. Wahdan, and S. M. Nassar, "Automated vision system for localizing structural defects in textile fabrics," Pattern Recogn. Lett. 26, 1435-1443 (2005).
[CrossRef]

Acharyya, M.

M. Acharyya and M. K. Kundu, "An adaptive approach to unsupervised texture segmentation using m-band wavelet transform," Signal Process. 81, 1337-1356 (2001).
[CrossRef]

Anagnostopoulos, C.

C. Anagnostopoulos, D. Vergados, E. Kayafas, V. Loumos, and G. Stassinopoulos, "A computer vision approach for textile quality control," J. Vis. Comput. Animat. 12, 31-44 (2001).
[CrossRef]

Attali, S.

F. S. Cohen, Z. Fan, and S. Attali, "Automated inspection of textile fabrics using textural models," IEEE Trans. Pattern Anal. Mach. Intell. 13, 803-808 (1991).
[CrossRef]

Bennamoun, M.

A. Bodnarova, M. Bennamoun, and S. Latham, "Optimal Gabor filters for textile flaw detection," Pattern Recogn. 35, 2973-2991 (2002).
[CrossRef]

A. Bodnarova, M. Bennamoun, and K. K. Kubik, "Suitability analysis of techniques for flaw detection in textiles using texture analysis," Pattern Anal. Appl. 3, 254-266 (2000).
[CrossRef]

Bodnarova, A.

A. Bodnarova, M. Bennamoun, and S. Latham, "Optimal Gabor filters for textile flaw detection," Pattern Recogn. 35, 2973-2991 (2002).
[CrossRef]

A. Bodnarova, M. Bennamoun, and K. K. Kubik, "Suitability analysis of techniques for flaw detection in textiles using texture analysis," Pattern Anal. Appl. 3, 254-266 (2000).
[CrossRef]

Chan, C. H.

C. H. Chan and G. K. H. Pang, "Fabric defect detection by Fourier analysis," IEEE Trans. Ind. Appl. 36, 1267-1276 (2000).
[CrossRef]

Chang, T.

T. Chang and C. C. J. Kuo, "Texture analysis and classification with tree-structured wavelet transform," IEEE Trans. Image Process. 2, 429-441 (1993).
[CrossRef] [PubMed]

Chen, C. F.

C. S. Lu, P. C. Chung, and C. F. Chen, "Unsupervised texture segmentation via wavelet transform," Pattern Recogn. 30, 729-742 (1997).
[CrossRef]

Chen, Y. H.

C. L. Tseng, Y. H. Chen, Y. Y. Xu, H. T. Pao, and H. C. Fu, "A self-growing probabilistic decision-based neural network with automatic data clustering," Ind. Math. 61, 21-38 (2004).

Chung, P. C.

C. S. Lu, P. C. Chung, and C. F. Chen, "Unsupervised texture segmentation via wavelet transform," Pattern Recogn. 30, 729-742 (1997).
[CrossRef]

Cohen, F. S.

F. S. Cohen, Z. Fan, and S. Attali, "Automated inspection of textile fabrics using textural models," IEEE Trans. Pattern Anal. Mach. Intell. 13, 803-808 (1991).
[CrossRef]

Cossu, R.

R. Cossu, I. H. Jermyn, and J. Zerubia, "Texture analysis using probabilistic models of the unimodal and multimodal statistics of adaptive wavelet packet coefficients," in Proceedings of IEEE 2004 International Conference on Acoustics, Speech, and Signal Processing (IEEE, 2004), pp. 597-600.

Dempster, A. P.

A. P. Dempster, N. M. Laird, and D. B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. R. Stat. Soc. Ser. B. Methodol. 39, 1-38 (1977).

Don, H. S.

H. S. Don, K. S. Fu, C. R. Liu, and W. C. Lin, "Metal surface inspection using image processing techniques," IEEE Trans. Syst. Man Cybern. 3, 139-146 (1984).

Eldeeb, H.

A. Abouelela, H. M. Abbas, H. Eldeeb, A. A. Wahdan, and S. M. Nassar, "Automated vision system for localizing structural defects in textile fabrics," Pattern Recogn. Lett. 26, 1435-1443 (2005).
[CrossRef]

Escofet, J.

J. Escofet, R. Navarro, M. S. Millan, and J. Pladellorens, "Detection of local defects in textiles webs using Gabor filters," Opt. Eng. (Bellingham) 37, 2297-2307 (1998).
[CrossRef]

Fan, J.

A. Laine and J. Fan, "Texture classification by wavelet packet signatures," IEEE Trans. Pattern Anal. Mach. Intell. 15, 1186-1191 (1993).
[CrossRef]

Fan, Z.

F. S. Cohen, Z. Fan, and S. Attali, "Automated inspection of textile fabrics using textural models," IEEE Trans. Pattern Anal. Mach. Intell. 13, 803-808 (1991).
[CrossRef]

Fu, H. C.

C. L. Tseng, Y. H. Chen, Y. Y. Xu, H. T. Pao, and H. C. Fu, "A self-growing probabilistic decision-based neural network with automatic data clustering," Ind. Math. 61, 21-38 (2004).

Fu, K. S.

H. S. Don, K. S. Fu, C. R. Liu, and W. C. Lin, "Metal surface inspection using image processing techniques," IEEE Trans. Syst. Man Cybern. 3, 139-146 (1984).

Furrokhnia, F.

A. K. Jain and F. Furrokhnia, "Unsupervised texture segmentation using Gabor filters," Pattern Recogn. 23, 1167-1186 (1991).
[CrossRef]

Goldberger, J.

J. Goldberger, S. Gordon, and H. Greenspan, "An efficient image similarity measure based on approximations of KL divergence between two Gaussian mixtures," in Proceedings of IEEE Ninth International Conference on Computer Vision (IEEE, 2003), pp. 487-493.
[CrossRef]

Gordon, S.

J. Goldberger, S. Gordon, and H. Greenspan, "An efficient image similarity measure based on approximations of KL divergence between two Gaussian mixtures," in Proceedings of IEEE Ninth International Conference on Computer Vision (IEEE, 2003), pp. 487-493.
[CrossRef]

Greenspan, H.

J. Goldberger, S. Gordon, and H. Greenspan, "An efficient image similarity measure based on approximations of KL divergence between two Gaussian mixtures," in Proceedings of IEEE Ninth International Conference on Computer Vision (IEEE, 2003), pp. 487-493.
[CrossRef]

Jain, A. K.

A. K. Jain and F. Furrokhnia, "Unsupervised texture segmentation using Gabor filters," Pattern Recogn. 23, 1167-1186 (1991).
[CrossRef]

Jasper, W. J.

W. J. Jasper and H. Potlapalli, "Image analysis of mispicks in woven fabric," Text. Res. J. 65, 683-692 (1995).
[CrossRef]

Jermyn, I. H.

R. Cossu, I. H. Jermyn, and J. Zerubia, "Texture analysis using probabilistic models of the unimodal and multimodal statistics of adaptive wavelet packet coefficients," in Proceedings of IEEE 2004 International Conference on Acoustics, Speech, and Signal Processing (IEEE, 2004), pp. 597-600.

Kang, T. J.

S. C. Kim and T. J. Kang, "Texture classification and segmentation using wavelet packet frame and Gaussian mixture model," Pattern Recogn. (to be published).

Kayafas, E.

C. Anagnostopoulos, D. Vergados, E. Kayafas, V. Loumos, and G. Stassinopoulos, "A computer vision approach for textile quality control," J. Vis. Comput. Animat. 12, 31-44 (2001).
[CrossRef]

Kim, S. C.

S. C. Kim and T. J. Kang, "Texture classification and segmentation using wavelet packet frame and Gaussian mixture model," Pattern Recogn. (to be published).

Kubik, K. K.

A. Bodnarova, M. Bennamoun, and K. K. Kubik, "Suitability analysis of techniques for flaw detection in textiles using texture analysis," Pattern Anal. Appl. 3, 254-266 (2000).
[CrossRef]

Kumar, A.

A. Kumar and G. K. H. Pang, "Defect detection in textured materials using Gabor filters," IEEE Trans. Ind. Appl. 38, 425-440 (2002).
[CrossRef]

A. Kumar and G. K. H. Pang, "Defect detection in textured materials using optimized filters," IEEE Trans. Syst. Man Cybern. Part B: Cybern. 32(5), 553-570 (2002).
[CrossRef]

Kundu, M. K.

M. Acharyya and M. K. Kundu, "An adaptive approach to unsupervised texture segmentation using m-band wavelet transform," Signal Process. 81, 1337-1356 (2001).
[CrossRef]

Kuo, C. C. J.

T. Chang and C. C. J. Kuo, "Texture analysis and classification with tree-structured wavelet transform," IEEE Trans. Image Process. 2, 429-441 (1993).
[CrossRef] [PubMed]

Laine, A.

A. Laine and J. Fan, "Texture classification by wavelet packet signatures," IEEE Trans. Pattern Anal. Mach. Intell. 15, 1186-1191 (1993).
[CrossRef]

Laird, N. M.

A. P. Dempster, N. M. Laird, and D. B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. R. Stat. Soc. Ser. B. Methodol. 39, 1-38 (1977).

Latham, S.

A. Bodnarova, M. Bennamoun, and S. Latham, "Optimal Gabor filters for textile flaw detection," Pattern Recogn. 35, 2973-2991 (2002).
[CrossRef]

Li, C. K.

K. K. Yiu, M. W. Mak, and C. K. Li, "Gaussian mixture models and probabilistic decision-based neural networks for pattern classification: a comparative study," Neural Comput. Appl. 8, 235-245 (1999).
[CrossRef]

Lin, W. C.

H. S. Don, K. S. Fu, C. R. Liu, and W. C. Lin, "Metal surface inspection using image processing techniques," IEEE Trans. Syst. Man Cybern. 3, 139-146 (1984).

Liu, C. R.

H. S. Don, K. S. Fu, C. R. Liu, and W. C. Lin, "Metal surface inspection using image processing techniques," IEEE Trans. Syst. Man Cybern. 3, 139-146 (1984).

Loumos, V.

C. Anagnostopoulos, D. Vergados, E. Kayafas, V. Loumos, and G. Stassinopoulos, "A computer vision approach for textile quality control," J. Vis. Comput. Animat. 12, 31-44 (2001).
[CrossRef]

Lu, C. S.

C. S. Lu, P. C. Chung, and C. F. Chen, "Unsupervised texture segmentation via wavelet transform," Pattern Recogn. 30, 729-742 (1997).
[CrossRef]

Mak, M. W.

K. K. Yiu, M. W. Mak, and C. K. Li, "Gaussian mixture models and probabilistic decision-based neural networks for pattern classification: a comparative study," Neural Comput. Appl. 8, 235-245 (1999).
[CrossRef]

Millan, M. S.

J. Escofet, R. Navarro, M. S. Millan, and J. Pladellorens, "Detection of local defects in textiles webs using Gabor filters," Opt. Eng. (Bellingham) 37, 2297-2307 (1998).
[CrossRef]

Nassar, S. M.

A. Abouelela, H. M. Abbas, H. Eldeeb, A. A. Wahdan, and S. M. Nassar, "Automated vision system for localizing structural defects in textile fabrics," Pattern Recogn. Lett. 26, 1435-1443 (2005).
[CrossRef]

Navarro, R.

J. Escofet, R. Navarro, M. S. Millan, and J. Pladellorens, "Detection of local defects in textiles webs using Gabor filters," Opt. Eng. (Bellingham) 37, 2297-2307 (1998).
[CrossRef]

Pang, G. K. H.

A. Kumar and G. K. H. Pang, "Defect detection in textured materials using Gabor filters," IEEE Trans. Ind. Appl. 38, 425-440 (2002).
[CrossRef]

A. Kumar and G. K. H. Pang, "Defect detection in textured materials using optimized filters," IEEE Trans. Syst. Man Cybern. Part B: Cybern. 32(5), 553-570 (2002).
[CrossRef]

X. Yang, G. K. H. Pang, and N. Yung, "Discriminative fabric defect detection using adaptive wavelet," Opt. Eng. (Bellingham) 41, 3116-3226 (2002).
[CrossRef]

C. H. Chan and G. K. H. Pang, "Fabric defect detection by Fourier analysis," IEEE Trans. Ind. Appl. 36, 1267-1276 (2000).
[CrossRef]

Pao, H. T.

C. L. Tseng, Y. H. Chen, Y. Y. Xu, H. T. Pao, and H. C. Fu, "A self-growing probabilistic decision-based neural network with automatic data clustering," Ind. Math. 61, 21-38 (2004).

Pladellorens, J.

J. Escofet, R. Navarro, M. S. Millan, and J. Pladellorens, "Detection of local defects in textiles webs using Gabor filters," Opt. Eng. (Bellingham) 37, 2297-2307 (1998).
[CrossRef]

Potlapalli, H.

W. J. Jasper and H. Potlapalli, "Image analysis of mispicks in woven fabric," Text. Res. J. 65, 683-692 (1995).
[CrossRef]

Rubin, D. B.

A. P. Dempster, N. M. Laird, and D. B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. R. Stat. Soc. Ser. B. Methodol. 39, 1-38 (1977).

Scheunders, P.

G. Van de Wouwer, P. Scheunders, and D. Van Dyck, "Statistical texture characterization from discrete wavelet representations," IEEE Trans. Image Process. 8, 592-598 (1999).
[CrossRef]

Schicktanz, K.

K. Schicktanz, "Automatic fault detection possibilities on nonwoven fabrics," Melliand Textilber. 74, 294-295 (1993).

Stassinopoulos, G.

C. Anagnostopoulos, D. Vergados, E. Kayafas, V. Loumos, and G. Stassinopoulos, "A computer vision approach for textile quality control," J. Vis. Comput. Animat. 12, 31-44 (2001).
[CrossRef]

Tseng, C. L.

C. L. Tseng, Y. H. Chen, Y. Y. Xu, H. T. Pao, and H. C. Fu, "A self-growing probabilistic decision-based neural network with automatic data clustering," Ind. Math. 61, 21-38 (2004).

Unser, M.

M. Unser, "Texture classification and segmentation using wavelet frames," IEEE Trans. Image Process. 4, 1549-1560 (1995).
[CrossRef] [PubMed]

Van de Wouwer, G.

G. Van de Wouwer, P. Scheunders, and D. Van Dyck, "Statistical texture characterization from discrete wavelet representations," IEEE Trans. Image Process. 8, 592-598 (1999).
[CrossRef]

Van Dyck, D.

G. Van de Wouwer, P. Scheunders, and D. Van Dyck, "Statistical texture characterization from discrete wavelet representations," IEEE Trans. Image Process. 8, 592-598 (1999).
[CrossRef]

Vergados, D.

C. Anagnostopoulos, D. Vergados, E. Kayafas, V. Loumos, and G. Stassinopoulos, "A computer vision approach for textile quality control," J. Vis. Comput. Animat. 12, 31-44 (2001).
[CrossRef]

Wahdan, A. A.

A. Abouelela, H. M. Abbas, H. Eldeeb, A. A. Wahdan, and S. M. Nassar, "Automated vision system for localizing structural defects in textile fabrics," Pattern Recogn. Lett. 26, 1435-1443 (2005).
[CrossRef]

Wood, E. J.

E. J. Wood, "Applying Fourier and associated transforms to pattern characterization in textile," Text. Res. J. 60, 212-220 (1990).
[CrossRef]

Xu, B.

B. Xu, "Identifying fabric structure with fast Fourier transform techniques," Text. Res. J. 66, 496-506 (1996).
[CrossRef]

Xu, Y. Y.

C. L. Tseng, Y. H. Chen, Y. Y. Xu, H. T. Pao, and H. C. Fu, "A self-growing probabilistic decision-based neural network with automatic data clustering," Ind. Math. 61, 21-38 (2004).

Yang, X.

X. Yang, G. K. H. Pang, and N. Yung, "Discriminative fabric defect detection using adaptive wavelet," Opt. Eng. (Bellingham) 41, 3116-3226 (2002).
[CrossRef]

Yiu, K. K.

K. K. Yiu, M. W. Mak, and C. K. Li, "Gaussian mixture models and probabilistic decision-based neural networks for pattern classification: a comparative study," Neural Comput. Appl. 8, 235-245 (1999).
[CrossRef]

Yung, N.

X. Yang, G. K. H. Pang, and N. Yung, "Discriminative fabric defect detection using adaptive wavelet," Opt. Eng. (Bellingham) 41, 3116-3226 (2002).
[CrossRef]

Zerubia, J.

R. Cossu, I. H. Jermyn, and J. Zerubia, "Texture analysis using probabilistic models of the unimodal and multimodal statistics of adaptive wavelet packet coefficients," in Proceedings of IEEE 2004 International Conference on Acoustics, Speech, and Signal Processing (IEEE, 2004), pp. 597-600.

IEEE Trans. Image Process. (3)

T. Chang and C. C. J. Kuo, "Texture analysis and classification with tree-structured wavelet transform," IEEE Trans. Image Process. 2, 429-441 (1993).
[CrossRef] [PubMed]

M. Unser, "Texture classification and segmentation using wavelet frames," IEEE Trans. Image Process. 4, 1549-1560 (1995).
[CrossRef] [PubMed]

G. Van de Wouwer, P. Scheunders, and D. Van Dyck, "Statistical texture characterization from discrete wavelet representations," IEEE Trans. Image Process. 8, 592-598 (1999).
[CrossRef]

IEEE Trans. Ind. Appl. (2)

C. H. Chan and G. K. H. Pang, "Fabric defect detection by Fourier analysis," IEEE Trans. Ind. Appl. 36, 1267-1276 (2000).
[CrossRef]

A. Kumar and G. K. H. Pang, "Defect detection in textured materials using Gabor filters," IEEE Trans. Ind. Appl. 38, 425-440 (2002).
[CrossRef]

IEEE Trans. Pattern Anal. Mach. Intell. (2)

F. S. Cohen, Z. Fan, and S. Attali, "Automated inspection of textile fabrics using textural models," IEEE Trans. Pattern Anal. Mach. Intell. 13, 803-808 (1991).
[CrossRef]

A. Laine and J. Fan, "Texture classification by wavelet packet signatures," IEEE Trans. Pattern Anal. Mach. Intell. 15, 1186-1191 (1993).
[CrossRef]

IEEE Trans. Syst. Man Cybern. (1)

H. S. Don, K. S. Fu, C. R. Liu, and W. C. Lin, "Metal surface inspection using image processing techniques," IEEE Trans. Syst. Man Cybern. 3, 139-146 (1984).

IEEE Trans. Syst. Man Cybern. Part B: Cybern. (1)

A. Kumar and G. K. H. Pang, "Defect detection in textured materials using optimized filters," IEEE Trans. Syst. Man Cybern. Part B: Cybern. 32(5), 553-570 (2002).
[CrossRef]

Ind. Math. (1)

C. L. Tseng, Y. H. Chen, Y. Y. Xu, H. T. Pao, and H. C. Fu, "A self-growing probabilistic decision-based neural network with automatic data clustering," Ind. Math. 61, 21-38 (2004).

J. R. Stat. Soc. Ser. B. Methodol. (1)

A. P. Dempster, N. M. Laird, and D. B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. R. Stat. Soc. Ser. B. Methodol. 39, 1-38 (1977).

J. Vis. Comput. Animat. (1)

C. Anagnostopoulos, D. Vergados, E. Kayafas, V. Loumos, and G. Stassinopoulos, "A computer vision approach for textile quality control," J. Vis. Comput. Animat. 12, 31-44 (2001).
[CrossRef]

Melliand Textilber. (1)

K. Schicktanz, "Automatic fault detection possibilities on nonwoven fabrics," Melliand Textilber. 74, 294-295 (1993).

Neural Comput. Appl. (1)

K. K. Yiu, M. W. Mak, and C. K. Li, "Gaussian mixture models and probabilistic decision-based neural networks for pattern classification: a comparative study," Neural Comput. Appl. 8, 235-245 (1999).
[CrossRef]

Opt. Eng. (Bellingham) (2)

J. Escofet, R. Navarro, M. S. Millan, and J. Pladellorens, "Detection of local defects in textiles webs using Gabor filters," Opt. Eng. (Bellingham) 37, 2297-2307 (1998).
[CrossRef]

X. Yang, G. K. H. Pang, and N. Yung, "Discriminative fabric defect detection using adaptive wavelet," Opt. Eng. (Bellingham) 41, 3116-3226 (2002).
[CrossRef]

Pattern Anal. Appl. (1)

A. Bodnarova, M. Bennamoun, and K. K. Kubik, "Suitability analysis of techniques for flaw detection in textiles using texture analysis," Pattern Anal. Appl. 3, 254-266 (2000).
[CrossRef]

Pattern Recogn. (3)

A. K. Jain and F. Furrokhnia, "Unsupervised texture segmentation using Gabor filters," Pattern Recogn. 23, 1167-1186 (1991).
[CrossRef]

A. Bodnarova, M. Bennamoun, and S. Latham, "Optimal Gabor filters for textile flaw detection," Pattern Recogn. 35, 2973-2991 (2002).
[CrossRef]

C. S. Lu, P. C. Chung, and C. F. Chen, "Unsupervised texture segmentation via wavelet transform," Pattern Recogn. 30, 729-742 (1997).
[CrossRef]

Pattern Recogn. Lett. (1)

A. Abouelela, H. M. Abbas, H. Eldeeb, A. A. Wahdan, and S. M. Nassar, "Automated vision system for localizing structural defects in textile fabrics," Pattern Recogn. Lett. 26, 1435-1443 (2005).
[CrossRef]

Signal Process. (1)

M. Acharyya and M. K. Kundu, "An adaptive approach to unsupervised texture segmentation using m-band wavelet transform," Signal Process. 81, 1337-1356 (2001).
[CrossRef]

Text. Res. J. (3)

W. J. Jasper and H. Potlapalli, "Image analysis of mispicks in woven fabric," Text. Res. J. 65, 683-692 (1995).
[CrossRef]

E. J. Wood, "Applying Fourier and associated transforms to pattern characterization in textile," Text. Res. J. 60, 212-220 (1990).
[CrossRef]

B. Xu, "Identifying fabric structure with fast Fourier transform techniques," Text. Res. J. 66, 496-506 (1996).
[CrossRef]

Other (3)

J. Goldberger, S. Gordon, and H. Greenspan, "An efficient image similarity measure based on approximations of KL divergence between two Gaussian mixtures," in Proceedings of IEEE Ninth International Conference on Computer Vision (IEEE, 2003), pp. 487-493.
[CrossRef]

S. C. Kim and T. J. Kang, "Texture classification and segmentation using wavelet packet frame and Gaussian mixture model," Pattern Recogn. (to be published).

R. Cossu, I. H. Jermyn, and J. Zerubia, "Texture analysis using probabilistic models of the unimodal and multimodal statistics of adaptive wavelet packet coefficients," in Proceedings of IEEE 2004 International Conference on Acoustics, Speech, and Signal Processing (IEEE, 2004), pp. 597-600.

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (12)

Fig. 1
Fig. 1

Example of subband histograms (solid curves) and fitted models (dashed curves) by the three-component GMM for a defect-free fabric image. (a) Fabric image with background texture. (b), (c), (d), (e) Approximation, horizontal, vertical, and diagonal detail images, respectively, at decomposition level 1. (f), (g), (h), (i) Subband histograms and fitted models of (b), (c), (d), and (e), respectively.

Fig. 2
Fig. 2

Diagram of feature extraction method using WPFD.

Fig. 3
Fig. 3

Schematic diagram for fabric defect detection using GMM.

Fig. 4
Fig. 4

Histograms (solid curves) of the KL-transformed WPFD coefficients corresponding to (a) the largest, (b) second largest, (c) third largest, and (d) fourth largest eigenvalues and fitted models (dashed curves) by the three-component GMM for Fig. 1a.

Fig. 5
Fig. 5

BIC values corresponding to the number of mixture components of the GMM to describe the fabric texture shown in Fig. 1a.

Fig. 6
Fig. 6

Schematic diagram for UDD.

Fig. 7
Fig. 7

Effect of the decomposition level of the WPFD on the defect segmentation results. (a) Defect samples: top to bottom, drawback, sloughed filling, thick place, and wild filling. (b), (c), (d) Segmentation results at decomposition levels 1, 2, 3, respectively.

Fig. 8
Fig. 8

Effect of the standard deviation of the Gaussian low-pass filter on the defect segmentation results. (a) Defect samples same as Fig. 7. (b), (c), and (d) Segmentation results with the standard deviation of two less than, equal to, and two more than that calculated by Eq. (17), respectively.

Fig. 9
Fig. 9

Effect of the control constant λ on the defect segmentation results. (a) Defect samples same as Fig. 7. (b), (c), (d) Segmentation results with λ = 2.5 , 3.5 , 4.5 , respectively.

Fig. 10
Fig. 10

Supervised defect segmentation results (bottom row in each three-row block) for the defect type of (a) broken pick, (b) coarse pick, (c) color spots, (d) double picks, (e) foreign fiber, (f) harness balk, (g) jerk-in, (h) kinky filling, (i) knot with halos, (j) knot, (k) loom waste, (l) mispick1, (m) mispick2, (n) oil spot, (o) overshot, (p) shed-split, (q) slack ends, (r) slub, (s) soil end, (t) warp float, (u) warp floats.

Fig. 11
Fig. 11

Effect of the size of subblock on the segmentation results in UDD. Defects in (a), (b), (c), (d) are the same as in the four rows of Fig. 7, respectively, top to bottom.

Fig. 12
Fig. 12

Unsupervised defect segmentation results (bottom row in each three-row block) for the defect types of (a) broken pick, (b) coarse pick, (c) color spots, (d) double picks, (e) foreign fiber, (f) harness balk, (g) jerk-in, (h) kinky filling, (i) knot with halos, (j) knot, (k) loom waste, (l) mispick1, (m) mispick2, (n) oil spot, (o) overshot, (p) shed-split, (q) slack ends, (r) slub, (s) soil end, (t) warp float, (u) warp floats.

Tables (1)

Tables Icon

Table 1 Comparison of Algorithms’ Overall Detection (D), Misdetection (M), and False Alarm (FA) Rates in Supervised Defect Detection

Equations (21)

Equations on this page are rendered with MathJax. Learn more.

2 ( j 1 ) 2 W 2 p ( 2 j 1 x n ) = k h k 2 n 2 j 2 W p ( 2 j x k ) ,
2 ( j 1 ) 2 W 2 p + 1 ( 2 j 1 x n ) = k g k 2 n 2 j 2 W p ( 2 j x k ) ,
c 2 p , n j 1 = k h k 2 n c p , k j ,
c 2 p + 1 , n j 1 = k g k 2 n c p , k j ,
h j 1 , k = [ h ] 2 N j h j , k , g j 1 , k = [ g ] 2 N j g j , k ,
c 2 p , n j 1 = k h j 1 , k n c p , k j , c 2 p + 1 , n j 1 = k g j 1 , k n c p , k j .
p ( x ( t ) ) = r = 1 R P ( Θ r ) p ( x ( t ) Θ r ) ,
ϕ ( x ( t ) , w ) = log p ( x ( t ) ) = log [ r = 1 R P ( Θ r ) p ( x ( t ) Θ r ) ] ,
l ( w ; X ) = t = 1 N log [ r = 1 R P ( Θ r ) p ( x ( t ) Θ r ) ] .
F M X = log p ( x ( t ) ) l ( w ; X ) ,
S H X = n ( x ( t ) ) t = 1 N n ( x ( t ) ) ,
BIC ( M k , X ) t = 1 N [ 2 log p ( x ( t ) M k ) ] d ( M k ) log N ,
I f = K T I ,
I s q i ( x , y ) = I f i ( x , y ) 2 ( i = 1 , 2 , , L ) .
I s m i ( x , y ) = g ( x , y ) I s q i ( x , y ) ( i = 1 , 2 , , L ) ,
g ( x , y ) = 1 2 π σ exp ( x 2 + y 2 2 σ 2 ) ,
σ 1 2 2 u 0 ,
I i ( s ) = vec [ I s m i ( x , y ) ] .
LLD i = LL i m df ( i = 1 , 2 , , total number of pixels ) .
S ( x , y ) = [ 0 if LLD ( x , y ) T = λ σ df 1 otherwise ] ,
KL ( f g ) = i = 1 n α i log α i β i + i = 1 n α i { 1 2 [ log Σ 2 i Σ 1 i + Tr ( Σ 2 i 1 Σ 1 i ) + ( μ 1 i μ 2 i ) T Σ 2 i 1 ( μ 1 i μ 2 i ) ] } .

Metrics