Title :
Neural Network´s Classification of Terrestrial Surface by Spectral Measuring
Author :
Alla, Lavrenuk ; Lilia, Gnibeda ; Katherine, Yarova
Author_Institution :
Space Res. Inst., Nat. Acad. of Sci. of Ukraine, Kyiv
Abstract :
Intellectual method´s usage for spectral curves analysis is described in article. The method is based on neural networks. The Earth´s surface spectral characteristics estimation is described in this article. Experiments using neural networks were made with different parameters. Various learning methods were used. Neural networks are effective and useful for classification terrestrial surface by spectral measuring.
Keywords :
geophysics computing; learning (artificial intelligence); neural nets; pattern classification; remote sensing; spectral analysis; neural network; spectral curve analysis; spectral measurement; surface spectral characteristics estimation; terrestrial surface classification; Earth; Neural networks; Neurons; Pollution measurement; Shape; Snow; Software libraries; Space technology; Vegetation mapping; Water; Distance Earth exploration; neural networks;
Conference_Titel :
Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, 2005. IDAACS 2005. IEEE
Conference_Location :
Sofia
Print_ISBN :
0-7803-9445-3
Electronic_ISBN :
0-7803-9446-1
DOI :
10.1109/IDAACS.2005.282959