A comparative analysis of supervised land cover classification using multi-polarized PALSAR EM image fusion

Author

Mittel, Vikas ; Singh, D. ; Saini, L.M.

Author_Institution

ECE Deptt., N.I.T. Kurukshetra, Kurukshetra, India

fYear

2014

fDate

15-17 Dec. 2014

Firstpage

1

Lastpage

5

Abstract

Complementary information contained in each of the individual polarization mode of fully polarimetric PALSAR data is fused using Expectation Maximization (EM) algorithm to improve information contents in the fused image. A comparative analysis of supervised classifiers viz. parallelepiped and minimum distance is accomplished to assess the suitability of the particular classifier. It is also demonstrated that fusion indeed improves various figures of merit such as producer, user and overall accuracies.

Keywords

data analysis; expectation-maximisation algorithm; geophysical image processing; image fusion; land cover; pattern classification; synthetic aperture radar; classifier; complementary information; expectation maximization algorithm; individual polarization mode; land cover classification; merit figure; multipolarized PALSAR EM image fusion; polarimetric PALSAR data; producer; user; Accuracy; Buildings; Classification algorithms; Image fusion; Remote sensing; Scattering; Synthetic aperture radar; Expectation maximization method; PALSAR image data; land cover classification; multi-polarized image fusion; supervised classification algorithms;

fLanguage

English

Publisher

ieee

Conference_Titel

Industrial and Information Systems (ICIIS), 2014 9th International Conference on

Conference_Location

Gwalior

Print_ISBN

978-1-4799-6499-4

Type

conf

DOI

10.1109/ICIINFS.2014.7036514

Filename

7036514