№1, 2020

CLASSIFICATION OF INDIVIDUAL FLAT OBJECTS BY THE FOREL METHOD

Yaroslav Y. Kulkov, Sultan S. Sadykov, Arcady L. Zhiznyakov, Denis G. Privezentsev

With the growth of production automation, it became necessary to ensure the operation of many technological processes without human participation. These are mainly the processes associated with routine, repetitive work or dangerous to humans. Such processes include the classification of objects. Vision systems can be used to solve this problem. Today, visual control systems are widely in demand in various fields of science, industry and technology. In particular, they are used to obtain data on monitoring the state of objects, their location, and recognizing any objects of different forms. In this paper, it is proposed to test the applicability of the FOREL clustering algorithms for solving the problem of classifying individual flat objects based on their dimensionless features. The result of the work will be the probabilities of recognition of each object from a given test sample, on the basis of which it will be possible to draw a conclusion about the applicability of this method for solving the formulated problem (pp.16-21).

Keywords: FOREL, FOREL2, clustering, identification, dimensionless features.
DOI : 10.25045/jpit.v11.i1.02
References
  • Genkin V.L, Moskalev E.S. Recognition systems for automated production / V.L.Genkin, I.L.Erosh, L.: Mashinostroenie, Leningradskoe otdelenie, 1988, 246 p.
  • Panov S.S., Mazepin P.G., Smirnov V.A., Petrova L.N. Robotic assembly stands with technical vision and computer control systems // Assembly in mechanical engineering, instrument making, 2010, no 12., pp. 23–28.
  • Sadykov S.S., Kulkov Y.Y. Recognition of individual test and real flat objects based on the dimensionless marks of convex hulls // Telecommunication. 2017, no 3. pp. 28-35.
  • Zagoruyko N.G., Lbov G.S., Elkin N.V. Empirical patterns detection algorithms. Novosibirsk: Nauka, 1985, 112 p.
  • Sadykov S.S., Kulkov Y.Y. The study of the possibility of recognizing flat objects using the dimensionless marks of the convex hulls of their two-gradation images // Proceedings of the international symposium "Reliability and quality", 2017, pp.76–79.
  • Amsharuk E.I., Gorbachev T.F. FOREL digital image clustering // Kuzbass State University, 2014, 350 p.
  • Schekin V.P., Cherniy S.G., Borduk A.S. Adaptive fuzzy systems on forel class taxonomy // Software products and systems, 2014, 115 p.
  • Sadykov S.S., Kulkov Y.Y. Recognition of Flat Objects Based on Dimensionless Marks Based of Their Contours [Text] // CEUR Workshop Proceedings, 2017, vol.1940, pp.68–73.
  • Sadykov S.S., Kulkov Y.Y. Recognition of individual test and real flat objects based on the dimensionless marks of the contours of their binary images // Information Technology, 2016. vol 22, no 10, pp.747–753.