Classification of hyperdimensional data using data fusion approaches

Author

Benediktsson, Jon Atli ; Sveinsson, Johannes R.

Author_Institution

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

Volume

4

fYear

1997

fDate

3-8 Aug 1997

Firstpage

1669

Abstract

Statistical classification methods based on consensus from several data sources are considered with respect to classification and feature extraction of hyperdimensional 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. Decision boundary feature extraction is considered as a preprocessing method in the data fusion. Consensus theory optimized with neural networks outperforms all other methods in terms of test accuracies in the experiments

Keywords

feature extraction; geophysical signal processing; geophysical techniques; geophysics computing; image classification; neural nets; remote sensing; sensor fusion; combined classification; consensus; consensus theoretic methods; data fusion; decision boundary theory; feature extraction; geophysical measurement technique; hyperdimensional data; hyperspectral remote sensing; image classification; image processing; multispectral remote sensing; neural net; neural network; preprocessing; sensor fusion; statistical classification method; terrain mapping; weighting; weights; Councils; Covariance matrix; Data engineering; Feature extraction; Neural networks; Optical imaging; Optimization methods; Spectroscopy; Testing; Vectors;

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.1997.609016

Filename

609016