A multi-structural instrument for identifying surface defects on rails has been developed. The instrument operates based on intelligent analysis of video data, and it opens up a wide range of possibilities for using neural-network analysis of images in real time. The instrument can transform a color image into a zero-contrast image, normalize video images, and convert them to binary form. It is distinguished by its real-time noise suppression, evaluation of indicators that provide information on defects, and automatic neural-network classification of defects. The instrument is used together with a dynamic expert system that employs a production model of knowledge representation.
Similar content being viewed by others
References
Yu. F. Voronin, V. A. Kamaev, A. V. Matokhina, and S. A. Karpov, “Computer determination of a defect, its causes, and a method for eliminating it,” Lit. Proizv., No. 7, 17–24 (2004).
V. P. Samarai, V. P. Avdokushin, S. P. Doroshenko, and V. M. Maznyuk, “Computer prediction and diagnosis of casting defects based on the compaction of the mold,” Lit. Proizv., No. 9, 26–30 (2004).
V. A. Barkhatov, “Identification of defects with a special type of artificial neural network,” Defektoskopiya, No. 2, 28–39 (2006).
P. Tseraski, F. Lyukking, and Kh.-Yu. Raitsig, “In-line identification of defects in bars with automated inspection of the surface,” Cher. Metally, No. 9, 45–50 (1999).
V. V. Pravosudovich, V. P. Sokurenko, V. N. Danchenko, et al., Defects in Steel Ingots and Rolled Products, Intermet Inzhiniring, Moscow (2006).
This research was supported by Russian Federation Presidential Grant No. MK-4068.2015.8.
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated from Metallurg, No. 4, pp. 11–16, April, 2016.
Rights and permissions
About this article
Cite this article
Trofimov, V.B. Multi-Structural Instrument for Identifying Surface Defects on Rails. Metallurgist 60, 351–357 (2016). https://doi.org/10.1007/s11015-016-0298-3
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11015-016-0298-3