Polarimetric SAR Interferometry for Forest Analysis Based on the ESPRIT Algorithm

Hiroyoshi YAMADA  Yoshio YAMAGUCHI  Yunjin KIM  Ernesto RODRIGUEZ  Wolfgang-Martin BOERNER  

IEICE TRANSACTIONS on Electronics   Vol.E84-C   No.12   pp.1917-1924
Publication Date: 2001/12/01
Online ISSN: 
Print ISSN: 0916-8516
Type of Manuscript: Special Section PAPER (Special Issue on New Technologies in Signal Processing for Electromagnetic-wave Sensing and Imaging)
synthetic aperture radar,  interferometry,  polarimetry,  vegetation,  ESPRIT algorithm,  

Full Text: PDF(2.1MB)>>
Buy this Article

Synthetic aperture radar interferometry have been established in the past two decades, and used extensively for many applications including topographic mapping of terrain and surface deformation. Vegetation analysis is also a growing area of its application. In this paper, we propose an polarimetric SAR interferometry technique for interferometric phase extraction of each local scatterer. The estimated position of local scattering centers has an important information for effective tree height estimation of forest. The proposed method formulated for local scattering center extraction is based on the ESPRIT algorithm which is known for high-resolution capability of closely located incident waves. The method shows high-resolution performance when local scattered waves are uncorrelated and have different polarization characteristics. Using the method, the number of dominant local scattering centers and interferometric phases in each image pixel can be estimated directly. Validity of the algorithm is demonstrated by using examples derived from SIR-C data.