Title :
Statistical approach to the recognition and classification for remote sensing
Author :
Dumin, O. ; Koltunov, Y. ; Dumina, O. ; Katrich, V.
Author_Institution :
Sch. of Radio Phys., V.N. Karazin Kharkiv Nat. Univ., Kharkiv, Ukraine
Abstract :
The work contains mathematical models, numerical algorithms, as well as prospects for the development and application of a new method of pattern recognition and classification (PRC): classification based on probabilistic learning sample. To estimate the parameters of the distribution of probability of each class the EM-algorithm is used. Ways of the improving the proposed algorithm are presented.
Keywords :
geophysical techniques; geophysics computing; probability; remote sensing; statistical distributions; EM-algorithm; classification statistical approach; numerical algorithms; pattern classification method; pattern recognition method; probabilistic learning sample; probability distribution parameters; recognition statistical approach; remote sensing; Classification algorithms; Electromagnetics; Probability distribution; Remote sensing; Statistical analysis; Vectors;
Conference_Titel :
Mathematical Methods in Electromagnetic Theory (MMET), 2012 International Conference on
Conference_Location :
Kyiv
Print_ISBN :
978-1-4673-4478-4
DOI :
10.1109/MMET.2012.6331272
