JUSTIFICATION OF THE METHOD OF RADIAL BASIS FUNCTIONS FOR CLASSIFICATION OF GEOSPATIAL OBJECTS

Main Article Content

A. YARMOLENKO
А. KUTSAYEVA

Abstract

The development of new and more reliable methods of pattern recognition is an important issue nowadays. On the basis of the values of the radial basis functions a system of equations and the coefficients a dividing plane that extend the class classification of objects and to increase their stability is made. The equations have been made with the threshold as a free member or without. According to the results of the theoretical development the software package performing object classification, which consists of two interlinked software modules: oroi_data_corr24bitRBF1.pro and Sub Макрос1ENVI_RBF() is composed. The first is written in the algorithmic environment language IDL of ENVI and the second in the VISUAL BASIC environment in Excel. Studies have shown that the method of radial basis functions (RBF) when the recommended in this article the parameters unambiguously classify all objects.

Article Details

How to Cite
YARMOLENKO, A., & KUTSAYEVA А. (2017). JUSTIFICATION OF THE METHOD OF RADIAL BASIS FUNCTIONS FOR CLASSIFICATION OF GEOSPATIAL OBJECTS. Vestnik of Polotsk State University. Part F. Constructions. Applied Sciences, (8), 178-184. Retrieved from https://journals.psu.by/constructions/article/view/1331
Author Biography

A. YARMOLENKO, Новгородский государственный университет им. Ярослава Мудрого, Россия

д-р техн. наук, проф.

References

Галушкин, А.И. Нейронные сети: основы теории / А.И. Галушкин. – М. : Горячая линия, 2010. – 480 с.

Хайкин, С. Нейронные сети : полный курс / С. Хайкин. – 2-е изд. испр. ; пер. с англ. – М. : И.Д. Вильямс, 2006. – 1104 с.

Ярмоленко, А.С. Расчет взаимной информации между объектами для дешифрирования многоканальных изображений по ее максимуму / А.С. Ярмоленко, О.А. Куцаева // Изв. вузов. Геодезия и аэрофотосъемка. – 2016. – № 3 – С. 108–116.