Abstract
Decision support in equipment condition monitoring systems with image processing is analyzed. Long-run accumulation of information about earlier made decisions is used to realize the adaptiveness of the proposed approach. It is shown that unlike conventional classification problems, the recognition of abnormalities uses training samples supplemented with reward estimates of earlier decisions and can be tackled using reinforcement learning algorithms. We consider the basic stages of contextual multi-armed bandit algorithms during which the probabilistic distributions of each state are evaluated to evaluate the current knowledge of the states, and the decision space is explored to increase the decision-making efficiency. We propose a new decision-making method, which uses the probabilistic neural network to classify abnormal situation and the softmax rule to explore the decision space. A modelling experiment in image processing was carried out to show that our approach allows a higher accuracy of abnormality detection than other known methods, especially for small-size initial training samples.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Milov, V.R., Suslov, B.A., and Kryukov, O.V., Intellectual management decision support in gas industry, Autom. Remote Control, 2011, vol. 72, no. 5, pp. 1095–1101.
Huynh, K. T., Barros, A., and Berenguer, C., Maintenance decision-making for systems operating under indirect condition monitoring: Value of online information and impact of measurement uncertainty, IEEE Trans. Reliab., 2012, vol. 61, no. 2, pp. 410–425.
Savchenko, A.V., Clustering and maximum likelihood search for efficient statistical classification with mediumsized databases, Optimization Lett., 2015, pp. 1–13. doi 10.1007/s11590-015-0948-6. Date of access March 31, 2016.
Koutroumbas. K. and Theodoridis, S., Pattern Recognition, Boston: Academic Press, 2008, p. 840.
Audibert, J.Y., Munos, R., and Szepesvári, C., Exploration–exploitation tradeoff using variance estimates in multi-armed bandits, Theor. Comput. Sci., 2009, vol. 410, no. 19, pp. 1876–1902.
Sutton, R.S. and Barto, A.G., Reinforcement Learning: An Introduction, Cambridge: MIT Press, 1998, p. 322.
Van Hasselt, H., Reinforcement learning in continuous state and action spaces, in Reinforcement Learning, Springer Berlin Heidelberg, 2012, pp. 207–251.
Lu, T., Pál, D., and Pál, M., Contextual multi-armed bandits, Proc. of the International Conference on Artificial Intelligence and Statistics, 2010, pp. 485–492.
Li, L., Chu, W., Langford, J., and Schapire, R.E., A contextual-bandit approach to personalized news article recommendation, Proc. of the International Conference on World Wide Web, ACM, 2010, pp. 661–670.
Auer, P., Cesa-Bianchi, N., and Fischer, P., Finite-time analysis of the multiarmed bandit problem, Machine Learning, 2002, vol. 47, pp. 235–256.
Specht, D.F., A general regression neural network, IEEE Trans. Neural Networks, 1991, vol. 2, no. 6, pp. 568–576.
Sonka, M., Hlavac, V., and Boyle, R., Image Processing, Analysis, and Machine Vision, 4th ed., Cengage Learning, p. 872.
Specht, D.F., Probabilistic neural networks, Neural Networks, 1990, vol. 3, pp.109–118.
Aarts, E. and Korst, J., Simulated Annealing and Boltzmann Machines, Wiley, 1988, p. 284.
Chernousov, V.O. and Savchenko, A.V., A fast mathematical morphological algorithm of video-based moving forklift truck detection in noisy environment, Proc. of the International Conference on Analysis of Images, Social Networks, and Texts, CCIS, 2014, vol. 439, pp. 57–65.
Mathematical Morphology: From Theory to Applications, Najman, L. and Talbot, H., Eds., Wiley-ISTE, 2010, p. 507.
Shapiro, L. and Stockman, G., Computer Vision, Upper Saddle River, NJ: Prentice Hall, 2001, p. 752.
May, B.C., Korda, N., Lee, A., and Leslie, D.S., Optimistic Bayesian sampling in contextual-bandit problems, J. Machine Learning Res., 2012, vol. 13, no. 1, pp. 2069–2106.
Russell, S. and Norvig, P., Artificial Intelligence: A Modern Approach, 3 ed., Pearson, 2009, p. 1152.
Savchenko, A.V. and Belova, N.S., Statistical testing of segment homogeneity in classification of piecewise-regular objects, Int. J. Appl. Math. Comput. Sci., 2015, vol. 25, no. 4, pp. 915–925.
Savchenko, A.V., Probabilistic neural network with homogeneity testing in recognition of discrete patterns set, Neural Networks, 2013, vol. 46, pp. 227–241.
Li, Q. and Racine, J.S., Nonparametric Econometrics: Theory and Practice. Princeton University Press, 2007, p. 746.
Savchenko, A.V., Milov, V.R., Sevriukov, A.A., and Milov, D.V., Classification of the dangerous situations in the decision support systems to control the technical objects, Informatsyonno-Izmeritelnye i Upravlyaushiye sistemy, 2015, no. 11, pp. 52–58 (in Russian).
Haykin, S., Neural Networks and Learning Machines, 3rd ed., New Jersey: Prentice Hall, 2008, p. 936.
Savchenko, A.V., Adaptive video image recognition system using a committee machine, Opt. Mem. Neural Networks (Information Optics), 2012, vol. 21, no.4, pp. 219–226.
Xu, Y., Fang, G., Chen, S., Zou, J.J., and Ye, Z., Real-time image processing for vision-based weld seam tracking in robotic GMAW, Int. J. Adv. Manuf. Technol., 2014, vol. 73, no. 9–12, pp. 1413–425.
Diagnostics of gas tubes. http://gazdiagnoznarodru/gaz013html. Cited November 30, 2015.
Milov, V.R., Zaripova, J.H., and Derevyankin, A.V., The family of bayesian algorithms for structural and parametrical neural network classifiers synthesis, Proc. of the 9th International Conference on Pattern Recognition and Image Analysis, PRIA-9-2008, 2008, vol. 2, pp. 43–46
Author information
Authors and Affiliations
Corresponding author
About this article
Cite this article
Savchenko, A.V., Milov, V.R. The adaptive approach to abnormal situations recognition using images from condition monitoring systems. Opt. Mem. Neural Networks 25, 79–87 (2016). https://doi.org/10.3103/S1060992X16020090
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.3103/S1060992X16020090