Abstract
Defect classification is the key task of a steel surface defect detection system. The current defect classification algorithms have not taken the feature noise into consideration. In order to reduce the adverse impact of feature noise, an anti-noise multi-class classification method was proposed for steel surface defects. On the one hand, a novel anti-noise support vector hyper-spheres (ASVHs) classifier was formulated. For N types of defects, the ASVHs classifier built N hyper-spheres. These hyper-spheres were insensitive to feature and label noise. On the other hand, in order to reduce the costs of online time and storage space, the defect samples were pruned by support vector data description with parameter iteration adjustment strategy. In the end, the ASVHs classifier was built with sparse defect samples set and auxiliary information. Experimental results show that the novel multi-class classification method has high efficiency and accuracy for corrupted defect samples in steel surface.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
E. Amid, S.R. Aghdam, H. Amindavar, Int. J. Electr. Comput. Eng. 6 (2012) 693–697.
F. Dupont, C. Odet, M. Cartont, NDT&E Int. 30 (1997) 3–10.
M. Xiao, M.M. Jiang, G.Y. Li, L. Xie, L. Yi, EURASIP J. Image Vide. Process. 2017 (2017) 48.
S.Y. Tian, K. Xu, Metals 7 (2017) 311.
F.G. Bulnes, D.F. García, F. Javier De la Calle, R. Usamentiaga, J. Molleda, J. Nondestruct. Eval. 35 (2016) 54.
L. Yi, G.Y. Li, M.M. Jiang, Steel Res. Int. 88 (2017) 176–187.
S.Y. Zhou, Y.P. Chen, D.L. Zhang, J.M. Xie, Y.F. Zhou, Mater. Tehnol. 51 (2017) 123–131.
H.J. Hu, Y.X. Li, M.F. Liu, W.H. Liang, Multimed. Tools Appl. 69 (2014) 199–216.
D.Y. Cui, K.W. Xia, Math. Probl. Eng. 2017 (2017) 4257273.
M.X. Chu, R.F. Gong, A.N. Wang, ISIJ Int. 54 (2014)119–124.
M.X. Chu, X.P. Liu, R.F. Gong, L.M. Liu, Chemometr. Intell. Lab. Syst. 176 (2018) 108–118.
P.Y. Hao, J.H. Chiang, Y.H. Lin, Appl. Intell. 30 (2009) 98–111.
X.J. Peng, D. Xu, Neural Comput. Appl. 24 (2014) 1207–1220.
X.J. Peng, D. Xu, Inform. Sci. 221 (2013) 12–27.
Y.T. Xu, IEEE Trans. Cybern. 47 (2017) 1540–1550.
X.L. Huang, L. Shi, J.A.K. Suykens, IEEE Trans. Pattern Anal. Mach. Intell. 36 (2014) 984–997.
D.M.J. Tax, R.P.W. Duin, Mach. Learn. 54 (2004) 45–66.
M.X. Chu, A.N. Wang, R.F. Gong, M. Sha, ISIJ Int. 54 (2014) 1638–1645.
R.F. Gong, M.X. Chu, A.N. Wang, Y.H. Yang, ISIJ Int. 55 (2015) 207–212.
C.H. Lin, J.S. Tsai, C.T. Chiu, IEEE Trans. Image Process. 19 (2010) 2307–2320.
Z.Y. He, L.N. Sun, Appl. Opt. 54 (2015) 9823–9830.
M.H.J. Vala, A. Baxi, Int. J. Adv. Res. Comput. Eng. Technol. 2 (2013) 387–389.
H.J. Hu, Y. Liu, M.F. Liu, L.Q. Nie, Neurocomputer 181 (2016) 86–95.
M.X. Chu, R.F. Gong, S. Gao, J. Zhao, Chemometr. Intell. Lab. Syst. 171 (2017) 140–150.
C. Cortes, V. Vapnik, Mach. Learn. 20 (1995) 273–297.
R .Jayadeva, R. Khemchandani, S. Chandra, IEEE Trans. Pattern Anal. Mach. Intell. 29 (2007) 905–910.
B. Fei, J. Liu, IEEE Trans. Neural Netw. 17 (2006) 696–704.
Acknowledgements
The authors would like to thank the reviewers for the helpful comments and suggestions. This work was supported by the National Natural Science Foundation of China (No. 51674140), Natural Science Foundation of Liaoning Province, China (No. 20180550067), Department of Education of Liaoning Province, China (Nos. 2017LNQN11 and 2020LNZD06), University of Science and Technology Liaoning Talent Project Grants (No. 601011507-20) and University of Science and Technology Liaoning Team Building Grants (No. 601013360-17).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chu, Mx., Feng, Y., Yang, Yh. et al. Multi-class classification method for steel surface defects with feature noise. J. Iron Steel Res. Int. 28, 303–315 (2021). https://doi.org/10.1007/s42243-020-00501-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s42243-020-00501-1