350 rub
Journal Neurocomputers №6 for 2014 г.
Article in number:
Hybrid technologies in intelligent systems for medicines identification
Keywords:
hybrids structures
neural networks
with macrolayers
training sampling
decisions making
object identification
Authors:
R. А. Tomakova - Dr.Sc. (Eng.), Professor, Department of Software of Computing Machinery, South-West State University, Kursk, Russia. E-mail: tomakova@rambler.ru
S. А. Filist - Dr.Sc. (Eng.), Professor, Department of Software of Computing Machinery, South-West State University, Kursk, Russia. E-mail: SFilist@gmail.com
М. V. Tomakov - Ph.D. (Hist.), Professor, Department of Labour Protection and Environment, South-West State University, Kursk, Russia. E-mail: tomakov_m_v@rambler.ru
S. А. Filist - Dr.Sc. (Eng.), Professor, Department of Software of Computing Machinery, South-West State University, Kursk, Russia. E-mail: SFilist@gmail.com
М. V. Tomakov - Ph.D. (Hist.), Professor, Department of Labour Protection and Environment, South-West State University, Kursk, Russia. E-mail: tomakov_m_v@rambler.ru
Abstract:
There has proposed use of hybrid systems of recognition for objects. identification on the basis of its geometric characteristics under condition of different factors influence. Decisions making is carried out in every group of informative features with the following decisions aggregation. Hybrid approach to the identification has been realized on the basis of neural-net paradigm containing probabilistic neural networks with macrolayers.
Pages: 31,66-34,68
References
- R. А. Tomakova- Dr.Sc.(Eng.), Professor, Department of Software of ComputingMachinery, South-West State University, Kursk, Russia. E-mail:tomakova@rambler.ru
- S. А. Filist - Dr.Sc.(Eng.), Professor, Department of Software of Computing Machinery,South-West State University, Kursk, Russia. E-mail:SFilist@gmail.com
- М. V. Tomakov - Ph.D. (Hist.), Professor, Department of Labour Protection andEnvironment, South-West State University, Kursk, Russia. E-mail:tomakov_m_v@rambler.ru