Abstract
The article is concerned with the results of research of sulphur print images of transverse template of a continuous cast billet. It is believed that the reliability of information about the billet quality is quite low. Visual assessment of the image results in a subjective estimation, which is influenced greatly by human factor. The authors developed a scheme of control points distribution aimed at graphic information acquisition in the process of billet manufacture.
It was suggested to divide the images into three classes by the object (template) brightness-background ratio. In the course of algorithm sophistication the authors offered a cascade method of image classification. The method consists of image assessment by shape forming characteristics of the histogram, by the distance to the reference normalized histograms and on the basis of fuzzy logic methods.
The following results were obtained in the course of experimental operation of the cascade method: the simplified method making use of the shape forming characteristics uniquely identified 22% of all images, the simplified method taking into account the distance to the reference normalized histograms identified 70% of the remaining images and only fuzzy logic method made it possible to uniquely identify the rest of the images. The full classification can be achieved only when all cascades of the developed method are applied.
Keywords
Sulphur print images, image histogram, shape forming characteristics, distance between the objects, rules of object identification, cascade classification.
1. Logunova O.S., Matsko I.I., Posohov I.V., Luk’ynov S.I. Automatic system for intelligent support of continuous cast billet production control processes, The International Journal of Advanced Manufacturing Technology, 2014, no.74(9-12), pp. 1407-1418, doi: 10.1007/s00170-014-6056-4.
2. Matsko I.I. Snegirev Y.V., Logunova O.S. Data Acquisition and Preparation Methods for Continuously Cast Billets Quality Analysis Software, Applied Mechanics and Materials, 2012, no.110-116, pp. 3557-3562. Doi: 10.4028/ www.scientific.net/AMM.110-116.3557
3. Logunova O.S. Internal-defect formation and the thermal state of continuous-cast billet, Steel in Translation, 2008, no.38(10), pp. 849-852, doi: 10.3103/S0967091208100148.
4. Chapeiie O., Haffner P., Vapnik V. Support Vector Machines for Histogram-Based Image Classification, IEEE Transactions on neural networks, 1999, no. 10(5), pp. 1055 – 1064.
5. Foody, G. A relative Evaluation of MultiClass Image Classificafion by Suuport Vector Machines, IEEE Transactions on geoscience and remote sensing, 2008, no.42(6), pp.1335– 1343.
6. Michael J.L., Budynek J., Akamatsu S. Automatic Classification of Single Facial Images, IEEE Transactions on pattern analysis and machine intelligence, 1999, no.21(12), pp. 1357 – 1362.
7. Leemput K.V., Maes F., Vandermeulen D., Suetens P. Automated Model-Based Tissue Classification of MR Images of the Brain, IEEE Transactions on medical imaging, 1999, no.18(10), pp. 897 – 908.
8. Varma M. Zisserman A. A Statistical Approach to Texture Classification from Single Images, International Journal of Computer Vision, 2005, no.62(1/2), pp. 61–81.
9. Wang F. Fuzzy Supervised Classification of Remote SensingImages, IEEE Transactions on geoscience and remote sensing, 1990, no.28(2), pp. 194–201.
10. Wang F., Newkirk R. Design and implementation of knowledge-base system for remotely sensed change detection, J. Imaging Techn, 1987, no.13, pp. 116 – 122.
11. Cannon R.L., Dave J.V., Bezdek J.C., Kolsky H.G. Segmentation of a thematic Mapper Image using the fuzzy cmeans clustering algorithm, IEEE Trans. Geosci.Remote Sensing, GE-24, 1986, pp. 400–408.
12. Morlini I., Zani S. A New Class of Weighted Similarity Indices Using Polytomous Variables, Journal of Classification, 2012, no.29(2), pp. 199–226, doi: 10.1007 / s00357-012-9107-2.
13. Giordani P., Kiers H. FINDCLUS: Fuzzy INdividual Differences CLUStering, Journal of Classification, 2012, no.29(2), pp. 170–198, doi: 10.1007 / s00357-012-9109-0.
14. Contreras P., Murtagh F. Fast, Linear Time Hierarchical Clustering using the Baire Metric, Journal of Classification, 2012, № 29(2), pp. 118-143, doi: 10.1007 / s00357-012-9106-3.
15. Bar-Hen A., Gey S., Poggi J.-M.Influence Measures for CART Classification Trees, Journal of Classification, doi: 10.1007/s00357-015-9172-4.
16. Shestakov A.L., Sviridyuk G.A. O novoy kontseptsii belogo shuma [New Concept of White Noise], Obozrenie prikladnoy i promyshlennoy matematiki [Overview of Applied and Industrial Mathematics], 2012, no.19(2), pp. 287-288.
17. Shestakov A.L. Sviridyuk G.A., Khudyakov Yu.V. Dinamicheskie izmereniya v prostranstvakh shumov [Dynamic Measurements in the Region of “Noises”], Bulletin of the SouthUral State University. Series: Computer Technologies, Control, Radioelectronics, 2013, no. 13(2), pp. 4-11.
18. Poynton C. Rehabilitation of gamma, Photonics West'98 Electronic Imaging. International Society for Optics.
19. A Standard Default Color Space for the Internet – sRGB, M. Stokes, M. Anderson, S. Chandrasekar, R. Motta., 1996. URL: http://www.w3.org/Graphics/Color/sRGB.html.
20. Gonsales R., Woods R.Tsifrovaay obrabotka izobrazheniy [Digital Image Processing], Moscow: Tekhnosfera, 2005, 1104 p.
21. Shapiro L., Stockman J. Kompyuternoe zrenie [Computer Vision], Moscow, BINOM. Laboratory of knowledge, 2006, 752 p.
22. Prett Y. Tsifrovaay obrabotka izobrazheniy [Digital Image Processing], translated form English, Moscow, Mir, 1982, book 1, 312 p.
23. Otsu N. A Threshold Selection Method from GrayLevel Histograms, IEEE Transactions on Systems, Man, and Cybernetics, 1979, no.9(1), pp. 62-66.
24. Ridler T.W. and S. Calvard, Picture thresholding using an iterative selection method, IEEE Trans. System, Man and Cybernetics, SMC-8, 1978, pp. 630-632.
25. OST 14-1-236-91. Steel. Method of Macrostructure Control in Continuous Cast Billet for Bar Section and Tube Stock Production. In substitution of OST 14-4-73 in the part of Continuous Cast Billet, Intorduction, 1992–01–01, Moscow, Department of Metallurgy, USSR, 1991, 37 p.