Classification and integration of multitype data

Author

Benediktsson, Jon Atli ; Sveinsson, Johannes R. ; Arnason, Kolbeinn

Author_Institution

Eng. Res. Inst., Iceland Univ., Reykjavik, Iceland

Volume

1

fYear

1998

fDate

6-10 Jul 1998

Firstpage

177

Abstract

Neural network approaches and statistical classification methods based on consensus from several data sources are considered with respect to classification and integration of multitype data. The consensus theoretic methods need weighting mechanisms to control the influence of each data source in the combined classification. The weights are optimized in order to improve the combined classification accuracies. A nonlinear method which utilizes a neural network is used and trained on a feature reduced input set. This nonlinear method gives excellent results in experiments along with other neural network models

Keywords

geophysical signal processing; geophysical techniques; geophysics computing; image classification; neural nets; remote sensing; sensor fusion; combined classification; data integration; geophysical measurement technique; image classification; image processing; land surface; multitype data; neural net; neural network; nonlinear method; optimized weights; remote sensing; sensor fusion; statistical classification method; terrain mapping; weighting mechanism; Backpropagation algorithms; Clustering algorithms; Covariance matrix; Data engineering; Electronic mail; Iterative algorithms; Multi-layer neural network; Neural networks; Vector quantization; Weight control;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing Symposium Proceedings, 1998. IGARSS '98. 1998 IEEE International

Conference_Location

Seattle, WA

Print_ISBN

0-7803-4403-0

Type

conf

DOI

10.1109/IGARSS.1998.702844

Filename

702844