Abstract
Convolutional neural networks (CNNs) achieved impressive recognition rates in image classification tasks recently. In order to exploit those capabilities, we trained CNNs on a database of photometric stereo images of metal surface defects, i.e. rail defects. Those defects are cavities in the rail surface and are indication for further surface degradation right up to rail break. Due to security issues, defects have to be recognized early in order to take countermeasures in time. By means of differently colored light-sources illuminating the rail surfaces from different and constant directions, those cavities are made visible in a photometric dark-field setup. So far, a model-based approach has been used for image classification, which expressed the expected reflection properties of surface defects in contrast to non-defects. In this work, we experimented with classical CNNs trained in pure supervised manner and also explored the impact of regularization methods such as unsupervised layer-wise pre-training and training data-set augmentation. The classical CNN already distinctly outperforms the model-based approach. Moreover, regularization methods yet yield further improvements.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Ciresan, D., Meier, U., Schmidhuber, J.: Multi-column deep neural networks for image classification. In: Proc. of Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3642–3649 (June 2012)
Ciresan, D., Meier, U., Masci, J., Schmidhuber, J.: A committee of neural networks for traffic sign classification. In: Proc. of International Joint Conference on Neural Networks (IJCNN), pp. 1918–1921 (July 2011)
Goodfellow, I.J., Bulatov, Y., Ibarz, J., Arnoud, S., Shet, V.: Multi-digit number recognition from street view imagery using deep convolutional neural networks. In: Proc. of International Conference on Learning Representations (ICLR) April
Doherty, A., Clark, S., Care, R., Dembowsky, M.: Why rails crack. Ingenia (23), 23–28 (2005)
Huber-Mörk, R., Nölle, M., Oberhauser, A., Fischmeister, E.: Statistical rail surface classification based on 2D and 21/2D image analysis. In: Blanc-Talon, J., Bone, D., Philips, W., Popescu, D., Scheunders, P. (eds.) ACIVS 2010, Part I. LNCS, vol. 6474, pp. 50–61. Springer, Heidelberg (2010)
Soukup, D., Huber-Mörk, R.: Cross-channel co-occurrence matrices for robust characterization of surface disruptions in 21/2D rail image analysis. In: Blanc-Talon, J., Philips, W., Popescu, D., Scheunders, P., Zemčík, P. (eds.) ACIVS 2012. LNCS, vol. 7517, pp. 167–177. Springer, Heidelberg (2012)
Woodham, R.J.: Photometric method for determining surface orientation from multiple images. Optical Engineering 19(1), 139–144 (1980)
Basri, R., Jacobs, D., Kemelmacher, I.: Photometric stereo with general, unknown lighting. International Journal of Computer Vision 72(3), 239–257 (2007)
Arel, I., Rose, D.C., Karnowski, T.P.: Deep machine learning - a new frontier in artificial intelligence research. IEEE Computational Intelligence Magazine 5(4), 13–18 (2010)
Simard, P.Y., Steinkraus, D., Platt, J.C.: Best practices for convolutional neural networks applied to visual document analysis. In: Proc. of International Conference on Document Analysis and Recognition (ICDAR), pp. 958–963 (2003)
Coates, A., Lee, H., Ng, A.Y.: An analysis of single-layer networks in unsupervised feature learning. In: Proc. of International Conference on Artificial Intelligence and Statistics (AISTATS) (2011)
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proceedings of the IEEE 86(11), 2278–2324 (1998)
Fukushima, K.: Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological Cybernetics 36(4), 193–202 (1980)
Masci, J., Meier, U., Ciresan, D., Schmidhuber, J., Fricout, G.: Steel defect classification with max-pooling convolutional neural networks. In: Proc. of International Joint Conference on Neural Networks (IJCNN), pp. 1–6 (June 2012)
Westin, S.H., Li, H., Torrance, K.E.: A comparison of four BRDF models. In: Jensen, H.W., Keller, A. (eds.) Proc. of Eurographics Symposium on Rendering, pp. 1–10 (2004)
Herwig, J., Leßmann, S., Bürger, F., Pauli, J.: Adaptive anomaly detection within near-regular milling textures. In: Proc. International Symposium on Image and Signal Processing and Analysis, Trieste, Italy, pp. 106–111 (2013)
Pang, G.K.H., Chu, M.-H.: Automated optical inspection of solder paste based on 2.5D visual images. In: Proc. of International Conference on Mechatronics and Automation, pp. 982–987 (2009)
Hoßfeld, M., Chu, W., Adameck, M., Eich, M.: Fast 3D-vision system to classify metallic coins by their embossed topography. Electronic Letters on Computer Vision and Image Analysis 5(4), 47–63 (2006)
Ciresan, D.C., Masci, J., Meier, U., Schmidhuber, J.: Stacked convolutional auto-encoders for hierarchical feature extraction. In: Proc. of International Conference on Artificial Neural Networks, ICANN (2011)
Bookstein, F.L.: Principal warps: Thin plate splines and the decomposition of deformations. IEEE Trans. Pattern Anal. Mach. Intell. 11(6), 567–585 (1989)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Soukup, D., Huber-Mörk, R. (2014). Convolutional Neural Networks for Steel Surface Defect Detection from Photometric Stereo Images. In: Bebis, G., et al. Advances in Visual Computing. ISVC 2014. Lecture Notes in Computer Science, vol 8887. Springer, Cham. https://doi.org/10.1007/978-3-319-14249-4_64
Download citation
DOI: https://doi.org/10.1007/978-3-319-14249-4_64
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-14248-7
Online ISBN: 978-3-319-14249-4
eBook Packages: Computer ScienceComputer Science (R0)