Share Email Print
cover

Journal of Applied Remote Sensing

Integration of multidimensional parameters of polarimetric synthetic aperture radar images for land use and land cover classification
Author(s): Yingbao Yang; Shuang Yu; Yanwen Li; Dengsheng Lu
Format Member Price Non-Member Price
PDF $20.00 $25.00

Paper Abstract

Diverse parameters that are decomposed from quad polarimetric synthetic aperture radar (PolSAR) imagery become the important basis in the target recognition and classification. The selection of effective parameters is a very important research topic. This work aims to explore the algorithm of parameter selection based on the parametric statistics and multidimensional analysis. The proposed algorithm merges the parameters from different decomposed algorithms and the optimal parameters describing the backscattering characters of the targets are explored. The difference of parameters’ locations in three-dimensional spaces is the important basis of target differentiation. Based on the selected parameters, PolSAR images are classified using the object-oriented analysis and decision tree method. The experimental results indicate that the overall accuracy and Kappa coefficient of the classification using the integrated multidimensional parameters were higher than those using Freeman and H/A/α decomposed parameters. The advantage of this algorithm is to select optimal parameter combinations in multidimensional space by integrating many parameters from different decomposed algorithms.

Paper Details

Date Published: 27 November 2013
PDF: 17 pages
J. Appl. Remote Sens. 7(1) 073472 doi: 10.1117/1.JRS.7.073472
Published in: Journal of Applied Remote Sensing Volume 7, Issue 1
Show Author Affiliations
Yingbao Yang, Hohai Univ. (China)
Shuang Yu, Hohai Univ. (China)
Yanwen Li, National Marine Data and Information Service (China)
Dengsheng Lu, Michigan State Univ. (United States)


© SPIE. Terms of Use
Back to Top