Share Email Print

Journal of Applied Remote Sensing

Local incidence angle referenced classification on polarimetric synthetic aperture radar images in mountain glacier areas
Author(s): Lei Huang; Zhen Li; Bang-sen Tian
Format Member Price Non-Member Price
PDF $20.00 $25.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

The incidence angle is a crucial parameter for synthetic aperture radar (SAR). The local incidence angle differs greatly in mountainous areas, and it may lead to quite different backscatter properties for the same ground targets for different incidence angles. As a result, it has to be taken into consideration in image classification. This work demonstrates the importance of the local incidence angle and the necessity of considering this angle in SAR image classification in mountainous areas. A local incidence angle referenced method based on support vector machines is developed to perform classification. In this method, the datasets are divided into different zones according to local incidence angle, and the training and predicting are performed within these zones to eliminate the influence of the local incidence angles. The experiments are performed around the Dongkemadi Glacier in the central Qinghai-Tibetan plateau using two RADARSAT-2 polarimetric SAR images. This paper concentrates on the effect of the local incidence angles in SAR image classification. Compared to the method using only backscatter coefficients, the proposed method improves the overall classification accuracy by 6 to 8%. More important, it is verified that this method is more helpful for the ground cover types where the terrain changes sharply.

Paper Details

Date Published: 27 May 2016
PDF: 14 pages
J. Appl. Rem. Sens. 10(2) 025015 doi: 10.1117/1.JRS.10.025015
Published in: Journal of Applied Remote Sensing Volume 10, Issue 2
Show Author Affiliations
Lei Huang, Institute of Remote Sensing and Digital Earth (China)
Zhen Li, Institute of Remote Sensing and Digital Earth (China)
Bang-sen Tian, Institute of Remote Sensing and Digital Earth (China)

© SPIE. Terms of Use
Back to Top