Study on the characteristics of the supervised classification of remotely sensed data using artificial neural networks

Author

Paek, Kil N. ; Song, Young S. ; Chae, Hyo S. ; Kim, Kwang E.

Author_Institution

Min. & Miner. Res. Eng., Chonbuk Nat. Univ., Chonju, South Korea

Volume

1

fYear

1997

fDate

3-8 Aug 1997

Firstpage

528

Abstract

The characteristics of classification of remotely sensed data using artificial neural networks are investigated. The training method of neural networks consists of a generalized delta rule (GDR) and a conjugate gradient (CG). The GDR is divided into two methods, data adaptive and block adaptive. The effects of the number and order of input data and learning rate were analyzed in the training. Data adaptive and block adaptive methods showed similar trends of error convergence in the GDR. The CG especially with a small data set had faster error convergence than the GDR. The CG having low error in the training didn´t show good accuracy in the testing stage because of the overtraining effect

Keywords

adaptive signal processing; geophysical signal processing; geophysical techniques; geophysics computing; image classification; neural nets; remote sensing; adaptive signal processing; block adaptive; conjugate gradient; data adaptive; generalized delta rule; geophysical measurement technique; image classification; image processing; land surface; learning rate; neural net; neural network; remote sensing; supervised classification; terrain mapping; training method; Artificial neural networks; Character generation; Computer errors; Convergence; Data analysis; Minerals; Pattern recognition; Testing; Training data; Water resources;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing, 1997. IGARSS '97. Remote Sensing - A Scientific Vision for Sustainable Development., 1997 IEEE International

Print_ISBN

0-7803-3836-7

Type

conf

DOI

10.1109/IGARSS.1997.615933

Filename

615933