Image-based Object Classification of Defects in Steel using Data-driven Machine Learning Optimization

Fabian Bürger, Christoph Buck, Josef Pauli, Wolfram Luther

Abstract

In this paper we study the optimization process of an object classification task for an image-based steel quality measurement system. The goal is to distinguish hollow from solid defects inside of steel samples by using texture and shape features of reconstructed 3D objects. In order to optimize the classification results we propose a holistic machine learning framework that should automatically answer the question "How well do state-of-the-art machine learning methods work for my classification problem?" The framework consists of three layers, namely feature subset selection, feature transform and classifier which subsequently reduce the data dimensionality. A system configuration is defined by feature subset, feature transform function, classifier concept and corresponding parameters. In order to find the configuration with the highest classifier accuracies, the user only needs to provide a set of feature vectors and ground truth labels. The framework performs a totally data-driven optimization using partly heuristic grid search. We incorporate several popular machine learning concepts, such as Principal Component Analysis (PCA), Support Vector Machines (SVM) with different kernels, random trees and neural networks. We show that with our framework even non-experts can automatically generate a ready for use classifier system with a significantly higher accuracy compared to a manually arranged system.

References

  1. Bergstra, J. and Bengio, Y. (2012). Random search for hyper-parameter optimization. J. Mach. Learn. Res., 13(1):281-305.
  2. Beyer, K., Goldstein, J., Ramakrishnan, R., and Shaft, U. (1999). When is ”nearest neighbor” meaningful? In Beeri, C. and Buneman, P., editors, Database Theory ICDT99, volume 1540 of Lecture Notes in Computer Science, pages 217-235. Springer Berlin Heidelberg.
  3. Buck, C., Bürger, F., Herwig, J., and Thurau, M. (2013). Rapid inclusion and defect detection system for large steel volumes. ISIJ International, 53, No. 11. accepted.
  4. Bürger, F., Herwig, J., Thurau, M., Buck, C., Luther, W., and Pauli, J. (2013). An auto-adaptive measurement system for statistical modeling of non-metallic inclusions through image-based analysis of milled steel surfaces. In Bosse, H. and Schmitt, R., editors, ISMTII 2013, 11th International Symposium on Measurement Technology and Intelligent Instruments. Apprimus Wissenschaftsverlag.
  5. Chang, C.-C. and Lin, C.-J. (2011). LIBSVM: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology, 2:27:1-27:27.
  6. Doshi, N. and Schaefer, G. (2012). A comparative analysis of local binary pattern texture classification. In Visual Communications and Image Processing (VCIP), 2012 IEEE, pages 1-6.
  7. Falconer, K. (2003). Fractal geometry: mathematical foundations and applications. Wiley, 2 edition.
  8. Herwig, J., Buck, C., Thurau, M., Pauli, J., and Luther, W. (2012). Real-time characterization of non-metallic inclusions by optical scanning and milling of steel samples. In Proc. of SPIE Vol, volume 8430, pages 843010-1.
  9. Huang, C.-L. and Wang, C.-J. (2006). A GA-based feature selection and parameters optimizationfor support vector machines. Expert Systems with Applications, 31(2):231 - 240.
  10. Jain, A., Duin, R. P. W., and Mao, J. (2000). Statistical pattern recognition: a review. Pattern Analysis and Machine Intelligence, IEEE Transactions on, 22(1):4- 37.
  11. Juszczak, P., Tax, D., and Duin, R. (2002). Feature scaling in support vector data description. In Proc. ASCI, pages 95-102. Citeseer.
  12. Kohavi, R. and John, G. H. (1997). Wrappers for feature subset selection. Artificial Intelligence, 97(12):273 - 324.
  13. Lemke, C., Budka, M., and Gabrys, B. (2013). Metalearning: a survey of trends and technologies. Artificial Intelligence Review, pages 1-14.
  14. Lin, S.-W., Lee, Z.-J., Chen, S.-C., and Tseng, T.-Y. (2008a). Parameter determination of support vector machine and feature selection using simulated annealing approach. Applied Soft Computing, 8(4):1505 - 1512. Soft Computing for Dynamic Data Mining.
  15. Lin, S.-W., Ying, K.-C., Chen, S.-C., and Lee, Z.-J. (2008b). Particle swarm optimization for parameter determination and feature selection of support vector machines. Expert Systems with Applications, 35(4):1817 - 1824.
  16. Ohser, J. and Mücklich, F. (2000). Statistical analysis of microstructures in materials science. John Wiley New York.
  17. Reif, M., Shafait, F., Goldstein, M., Breuel, T., and Dengel, A. (2012). Automatic classifier selection for nonexperts. Pattern Analysis and Applications, pages 1- 14.
  18. Toriwaki, J. and Yoshida, H. (2009). Fundamentals of threedimensional digital image processing. Springer.
  19. Van der Maaten, L., Postma, E., and Van Den Herik, H. (2009). Dimensionality reduction: A comparative review. Journal of Machine Learning Research, 10:1- 41.
  20. Wolpert, D. H. (1996). The lack of a priori distinctions between learning algorithms. Neural computation, 8(7):1341-1390.
Download


Paper Citation


in Harvard Style

Bürger F., Buck C., Pauli J. and Luther W. (2014). Image-based Object Classification of Defects in Steel using Data-driven Machine Learning Optimization . In Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014) ISBN 978-989-758-004-8, pages 143-152. DOI: 10.5220/0004737101430152


in Bibtex Style

@conference{visapp14,
author={Fabian Bürger and Christoph Buck and Josef Pauli and Wolfram Luther},
title={Image-based Object Classification of Defects in Steel using Data-driven Machine Learning Optimization},
booktitle={Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014)},
year={2014},
pages={143-152},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004737101430152},
isbn={978-989-758-004-8},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 9th International Conference on Computer Vision Theory and Applications - Volume 2: VISAPP, (VISIGRAPP 2014)
TI - Image-based Object Classification of Defects in Steel using Data-driven Machine Learning Optimization
SN - 978-989-758-004-8
AU - Bürger F.
AU - Buck C.
AU - Pauli J.
AU - Luther W.
PY - 2014
SP - 143
EP - 152
DO - 10.5220/0004737101430152