350 rub
Journal Neurocomputers №8 for 2016 г.
Article in number:
Solution task of classification symbols by use neural network
Keywords: symbol recognition classification symbol neural network neural network training fuzzification rule base
Authors:
N.V. Andrievskaya - Ph. D. (Eng.), Associate Professor, Department of of Automation Means, Perm National Research Polytechnic University. E-mail: nataly-anv@mail.ru
Abstract:
Symbol recognition is one of the most difficult tasks. Algorithm of solution task of symbol recognition is described. The task of classification symbols is formulated. The neural network approach for solving the problem of classification symbols estimating is proposed. The schemes of neural network implementation and algorithms of neural network training are considered. Results of neural network simulation are described. . The analysis of the simulation results is
Pages: 10-13
References

 

  1. Demin A.A. Obzor intellektualnykh sistem dlja ocenki kalligrafii // Inzhenernyjj vestnik: ehlektronnyjj nauchno-tekhnicheskijj zhurnal. 2012. № 9. S. 1-25.
  2. Mesteckijj L.M. Matematicheskie metody raspoznavanija obrazov. M.: Izd-vo «INTUIT». 2008. 136 s.
  3. Nauck D., Nauck U., and Kruse R., NEFCLASS for JAVA - New Learning Algorithms. Seventh International Fuzzy Systems Association // World Congress IFSA \'97. Prague. 1997. P. 50-55.
  4. Andrievskaja N.V., Reznikov A.S., CHeranev A.A. Osobennosti primenenija nejjro-nechetkikh modelejj dlja zadach sinteza sistem avtomaticheskogo upravlenija // Fundamentalnye issledovanija. 2014. № 11-7. S. 1445-1449.
  5. Pen-Based Recognition of Handwritten Digits Data Set. [EHlektronnyjj resurs]. URL: http://archive.ics.uci.edu/ml/datasets/Pen-Based+Recognition+of+Handwritten+Digits.
  6. Simple NEFClass implementation in C# [EHlektronnyjj resurs]. URL: https://github.com/igor-kondratiev/NEFClassLib.