LANDSAT-TM image classification using principal components analysis and neural networks

Author

Sergi, R. ; Solaiman, B. ; Mouchot, M.C. ; Pasquariello, G. ; Posa, Pr

Author_Institution

Telecom Bretagne, Brest, France

Volume

3

fYear

34881

fDate

10-14 Jul1995

Firstpage

1927

Abstract

The application of three neural architectures in the classification LANDSAT images is conducted using multispectral data as well as the principal components projections. Results evaluation is given in terms of recognition rates and reconstructed images

Keywords

feedforward neural nets; geophysical signal processing; geophysical techniques; geophysics computing; image classification; multilayer perceptrons; optical information processing; remote sensing; self-organising feature maps; Kohonen neural net; LANDSAT; TM; feedforward neural net; geophysical signal processing; hybrid learning vector quantization; image classification; image processing; image recognition rate; infrared; land surface; measurement technique; multidimensional signal processing; multilayer perceptron; multispectral remote sensing; neural network; optical imaging; principal components analysis; principal components projections; reconstructed image; satellite remote sensing; self organizing feature map; terrain mapping; visible; Image classification; Image recognition; Image reconstruction; Neural networks; Organizing; Pattern recognition; Principal component analysis; Remote sensing; Satellites; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing Symposium, 1995. IGARSS '95. 'Quantitative Remote Sensing for Science and Applications', International

Conference_Location

Firenze

Print_ISBN

0-7803-2567-2

Type

conf

DOI

10.1109/IGARSS.1995.524069

Filename

524069