Skip to main content
SpringerLink
Log in

Imbalanced defect classification for mobile phone screen glass using multifractal features and a new sampling method

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

Defect classification has drawn significant attention in the mobile phone screen glass (MPSG) manufacturing field because it helps to determine problems in the manufacturing process. Two problems exist in MPSG defect classification: (1) the high dimensionality of the defect feature; (2) imbalanced defect example classification. The first problem tends to yield low accuracy for classifying overall defect examples, and the second problem has a low accuracy for minority ones. To address these two problems, an imbalanced MPSG defect classification scheme is presented. First, based on the multifractal spectrum, defect features are extracted to reduce the feature dimensionality. Defect features are distinguishably characterized by two multifractal metrics to promote the performance of classifying defects. Second, considering example contributions to determine the classification boundary, a new sampling method is proposed to address the imbalanced defect example classification. This method improves the classification accuracy of the minority class through implementation of different sampling strategies to SVs (support vectors) and NSVs (non support vectors) in the majority and minority classes. Experiments are conducted on real MPSG defect examples, and the experimental results show that the imbalanced MPSG defect classification scheme achieves a 96.61% overall accuracy and a 93.27% geometric mean of the classification accuracies of four-type defects; these results are superior to the results achieved by other methods used in the experiment.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

References

  1. Alibeigi M, Hashemi S, Hamzeh A (2012) DBFS: an effective density based feature selection scheme for small sample size and high dimensional imbalanced data sets[J]. Data Knowl Eng 81-82:67–103

    Article  Google Scholar 

  2. Barua S, Islam MM, Yao X, Murase K (2014) MWMOTE--majority weighted minority oversampling technique for imbalanced data set learning[J]. IEEE T Knowl Data En 26(2):405–425

    Article  Google Scholar 

  3. Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) SMOTE: synthetic minority over-sampling technique[J]. J Artif Intell Res 16(1):321–357

    MATH  Google Scholar 

  4. Chen Z, Shen Y, Bao W, Li P, Wang X, Ding Z (2015) Identification of surface defects on glass by parallel spectral domain optical coherence tomography[J]. Opt Express 23(18):23634–23646

    Article  Google Scholar 

  5. Cheng X, Chen Y, Tao Y, Wang C, Kim M, Lefcourt A (2004) A novel integrated PCA and FLD method on hyperspectral image feature extraction for cucumber chilling damage inspection[J]. Trans ASABE 47(4):1313–1320

    Article  Google Scholar 

  6. Cortes C, Vapnik V (1995) Support-vector networks[J]. Mach Learn 20(3):273–297

    MATH  Google Scholar 

  7. Drummond C, Holte RC (2003) C4. 5, class imbalance, and cost sensitivity: why under-sampling beats over-sampling. Workshop on learning from imbalanced datasets II, Citeseer

  8. Du JX, Zhai CM, Wang QP (2013) Recognition of plant leaf image based on fractal dimension features[J]. Neurocomputing 116(20):150–156

    Article  Google Scholar 

  9. Estabrooks A, Jo T, Japkowicz N (2004) A multiple resampling method for learning from imbalanced data sets[J]. Comput Intell 20(1):18–36

    Article  MathSciNet  Google Scholar 

  10. Farsaie Alaie H, Abou-Abbas L, Tadj C (2016) Cry-based infant pathology classification using GMMs[J]. Speech Comm 77:28–52

    Article  Google Scholar 

  11. Galar M, Fernandez A, Barrenechea E, Bustince H, Herrera F (2012) A review on ensembles for the class imbalance problem: bagging-, boosting-, and hybrid-based approaches[J]. IEEE T Syst Man Cy C 42(4):463–484

    Article  Google Scholar 

  12. Gokmen G (2014) The defect detection in glass materials by using discrete wavelet packet transform and artificial neural network[J]. J Vibroeng 16(3):1434–1443

    Google Scholar 

  13. Halsey TC, Jensen MH, Kadanoff LP, Procaccia I, Shraiman BI (1986) Fractal measures and their singularities: the characterization of strange sets[J]. Phys Rev A 33(2):1141–1151

    Article  MATH  MathSciNet  Google Scholar 

  14. Han H, Wang WY, Mao BH (2005) Borderline-SMOTE: a new over-sampling method in imbalanced data sets learning. Springer, Advances in intelligent computing, pp. 878–887

    Google Scholar 

  15. Hu H, Li Y, Liu M, Liang W (2014) Classification of defects in steel strip surface based on multiclass support vector machine[J]. Multimedia Tools and Applications 69(1):199–216

    Article  Google Scholar 

  16. Hu H, Liu Y, Liu M, Nie L (2016) Surface defect classification in large-scale strip steel image collection via hybrid chromosome genetic algorithm[J]. Neurocomputing 181:86–95

    Article  Google Scholar 

  17. Jian C, Gao J, Ao Y (2016a) Automatic surface defect detection for mobile phone screen glass based on machine vision[J]. Appl Soft Comput. doi:10.1016/j.asoc.2016.10.030

    Google Scholar 

  18. Jian CX, Gao J, Ao YH (2016b) A new sampling method for classifying imbalanced data based on support vector machine ensemble[J]. Neurocomputing 193(12):115–122

    Article  Google Scholar 

  19. Kotsiantis S, Kanellopoulos D, Pintelas P (2006) Handling imbalanced datasets: a review[J]. GESTS International Transactions on Computer Science and Engineering 30(1):25–36

    Google Scholar 

  20. Kubat M, Matwin S (1997) Addressing the curse of imbalanced training sets: one-sided selection. ICML, Nashville, USA, pp. 179–186

    Google Scholar 

  21. Li WC, Tsai DM (2012) Wavelet-based defect detection in solar wafer images with inhomogeneous texture[J]. Pattern Recogn 45(2):742–756

    Google Scholar 

  22. Li X, Wang L, Sung E (2008) AdaBoost with SVM-based component classifiers[J]. Eng Appl Artif Intell 21(5):785–795

    Article  Google Scholar 

  23. Li D, Liang LQ, Zhang WJ (2014) Defect inspection and extraction of the mobile phone cover glass based on the principal components analysis[J]. Int J Adv Manuf Tech 73(9–12):1605–1614

    Article  Google Scholar 

  24. Liang LQ, Li D, Fu X, Zhang WJ (2016) Touch screen defect inspection based on sparse representation in low resolution images[J]. Multimed Tools Appl 75(5):2655–2666

    Article  Google Scholar 

  25. Lin ZY, Hao ZF, Yang XW, Liu XA (2009) Several SVM ensemble methods integrated with under-sampling for imbalanced data learning. Springer, Advanced Data Mining and Applications, pp. 536–544

    Google Scholar 

  26. Liu YH, Liu YC (2009) Reducing the run-time complexity of support vector data descriptions, 2009 International Joint Conference on Neural Networks (IJCNN), (IEEE), pp 1863–1870

  27. Liu YH, Huang YK, Lee MJ (2008) Automatic inline defect detection for a thin film transistor–liquid crystal display array process using locally linear embedding and support vector data description[J]. Meas Sci Technol 19(9):095501

    Article  Google Scholar 

  28. Liu HG, Chen YP, Peng XQ, Xie JM (2011) A classification method of glass defect based on multiresolution and information fusion[J]. Int J Adv Manuf Tech 56(9–12):1079–1090

    Article  Google Scholar 

  29. Liu M, Liu Y, Hu H, Nie L (2016) Genetic algorithm and mathematical morphology based binarization method for strip steel defect image with non-uniform illumination[J]. J Vis Commun Image Represent 37:70–77

    Article  Google Scholar 

  30. Lopes R, Betrouni N (2009) Fractal and multifractal analysis: a review[J]. Med Image Anal 13(4):634–649

    Article  Google Scholar 

  31. Mandelbrot BB (1983) The fractal geometry of nature/revised and enlarged edition[J]. WH Freeman and Co., New York, p. 495

    Google Scholar 

  32. Nasira G, Banumathi P (2013) Fourier transform and image processing in automated fabric defect inspection system[J]. Int J Comput Int Info 3(1):61–64

    Google Scholar 

  33. Peng X, Chen Y, Yu W, Zhou Z, Sun G (2008) An online defects inspection method for float glass fabrication based on machine vision[J]. Int J Adv Manuf Tech 39(11–12):1180–1189

    Article  Google Scholar 

  34. Reed TR, Dubuf JH (1993) A review of recent texture segmentation and feature extraction techniques[J]. CVGIP: Image und 57(3):359–372

    Article  Google Scholar 

  35. Serdaroglu A, Ertuzun A, Erçil A (2006) Defect detection in textile fabric images using wavelet transforms and independent component analysis[J]. Pattern Recog Image Anal 16(1):61–64

    Article  Google Scholar 

  36. Sophian A, Tian GY, Taylor D, Rudlin J (2003) A feature extraction technique based on principal component analysis for pulsed Eddy current NDT[J]. NDT&E Int 36(1):37–41

    Article  Google Scholar 

  37. Tian G, Sophian A, Taylor D, Rudlin J (2005) Wavelet-based PCA defect classification and quantification for pulsed eddy current NDT, IEE Proceedings - Science, Measurement and Technology (IET2005), pp 141–148

  38. Tsai DM, Chuang ST (2009) 1D-based defect detection in patterned TFT-LCD panels using characteristic fractal dimension and correlations[J]. Mach Vision Appl 20(6):423–434

    Article  Google Scholar 

  39. Valavanis I, Kosmopoulos D (2010) Multiclass defect detection and classification in weld radiographic images using geometric and texture features[J]. Expert Syst Appl 37(12):7606–7614

    Article  Google Scholar 

  40. Veropoulos K, Campbell C, Cristianini N (1999) Controlling the sensitivity of support vector machines. Proceedings of the international joint conference on AI, pp 55–60

  41. Xu K, Ai YH, Wu XY (2013) Application of multi-scale feature extraction to surface defect classification of hot-rolled steels[J]. Int J Min Met Mater 20(1):37–41

    Article  Google Scholar 

Download references

Acknowledgments

This work was supported in part by the National Natural Science Foundation of China (Grant no.51675106, no.51275093), Guangdong Provincial Natural Science Foundation (Grant no.2015A030312008), and in part by the Guangdong Provincial R&D Key Projects (Grant no.2015B010104008, no.2016A030308016).

Author information

Authors and Affiliations

  1. Key Laboratory of Mechanical Equipment Manufacturing and Control Technology of Ministry of Education, Guangdong University of Technology, Guangzhou, 510006, China

    Chuanxia Jian & Jian Gao

  2. Guangdong Provincial Key Laboratory of Computer Integrated Manufacturing, Guangdong University of Technology, Guangzhou, 510006, China

    Yinhui Ao

Authors
  1. Chuanxia Jian
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jian Gao
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yinhui Ao
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Chuanxia Jian.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jian, C., Gao, J. & Ao, Y. Imbalanced defect classification for mobile phone screen glass using multifractal features and a new sampling method. Multimed Tools Appl 76, 24413–24434 (2017). https://doi.org/10.1007/s11042-016-4199-z

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-016-4199-z

Keywords

Search

Navigation