W.M. Boerner

Scene characterization using sub-aperture polarimetric SAR data analysis

In this paper is introduced a fully polarimetric sub-aperture analysis method. A deconvolution technique is developed in order to decompose synthesized SAR images into sub-aperture data sets which correspond to the scene global response observed under different azimuthal look angles. A polarimetric variation analysis is achieved, using pertinent parameters, to determine the nature of the non-stationary scattering mechanisms. A statistical analysis of the polarimetric parameters permits to clearly discriminate the media showing a varying behavior during the SAR integration. Decomposition and analysis techniques are applied to data acquired by the DLR airborne E-SAR sensor at L band.

Multi-frequency polarimetric SAR interferometry for vegetation structure extraction

The authors compare the relative information content of single baseline polarimetric interferometry at C-, L- and P-bands and conclude as to the best frequency range for vegetation structure estimation. They present an inversion algorithm for vegetated terrain based on a coherent polarimetric scattering model of volume and surface scattering. They point out some general aspects of using multi-frequency interferometric techniques for vegetation studies and use L- and P-band data from the DLR E-SAR system to verify the theoretical models.

Polarization Microwave Holography: An Extension Of Scalar To Vector Holography

Whereas in the optical region commonly catan hoeognaphy is used to describe the process of wavefront reconstruction, it is shown that polarization effects can no longer be neglected in the microwave and mm-wave regions and an extension to vegan hotogitaphy is required. Based on a succint review of previous studies on considering polarization properties in holography, it is apparent that a complete vector holographic treatment has yet to be formulated. Therefore, various basic properties of the object scattering matrices and their transformation invariants are introduced together with some pertinent descriptions of coherent versus non/incoherent scatter effects from rough surfaces which then allow a complete formulation of vector holography. Finally it is shown that a further generalization of vector holography incorporating recording of magnetic (energy) intensities in addition to the electric intensity, may enable us to recover simultaneously the shape and the averaged surface impedance of a scatterer relative to some reference surface impedance.

Interferometric phase and coherence of forest estimated by ESPRIT-based polarimetric SAR interferometry

Polarimetric SAR interferometry has been widely studied for forest observations. The technique utilizes polarization state difference among local scattering centers of the forest to decompose them. Using the method, we can estimate precise forest parameters. We proposed an alternative method based on the ESPRIT algorithm for the estimation. The method can detect as many as 3 local scatterers with fully polarimetric data, and can extract interferometric phase of each local scatterer. In this paper, we verify the availability of the method in 2- and 3-local-scatterer model for forest estimation, and show that forest analysis with 3-local-scatterer improve height estimation accuracy. SIR-C/X-SAR data were used for these verifications. In addition, experimental results of restricted dual-polarization data are also provided to show availability of the method.

Mapping land uplift and subsidence in the industrial parks in northern Taiwan by radar interferometry

The prevailing complex geological and ecological conditions of Taiwan have drawn considerable attention from various geo-ecological communities because of their vulnerability to produce various natural hazards at different scales. Located in the tropical/subtropical zone of the Pacific Rim, its ecological and rugged mountainous properties are environmentally sensitive making monitoring and observation especially difficult because of the high population density. In this article, we have investigated the land deformation in two adjacent industrial parks, Jhong-Li and Guei-Shan, in northern Taiwan using radar interferometry. The Interferometric Synthetic Aperture Radar technique for processing a series of data sets was first validated by comparison with ground levelling measurements over a test site. Excellent agreement was obtained in both deformation pattern and magnitude of subsidence rate. The period of observation dated from 1993 to 2000 with Synthetic Aperture Radar images from ERS-1 and ERS-2. The results, after least-squares adjustment, revealed that the maximum subsidence reached 10 cm and the subsidence rates were about 1.8 cm year−1 (at epicentres) since 1993 at both parks. It was also found that the subsidence rate slowed down after 1998 at Jhong-Li park while continuing at Guei-Shan park. This was strongly associated with local groundwater extraction activities.

Forest height feature extraction in polarimetric SAR interferometry by using rotational invariance property

This paper present a new forest height feature extraction technique for single baseline polarimetric and interferometric SAR data. The authors have proposed polarimetric SAR interferometry based on the ESPRIT algorithm. However, the algorithm assumes that there exist dominant polarized components in both the ground and canopy. Hence, estimated interferometric phase of local scattering centers may be slightly biased when forest components are highly depolarized. In this report, we examine effect of depolarized components of forest in ESPRIT-based polarimetric SAR interferometry. Numerical examples show that the effective scenario for the algorithm. Also, we present an alternative derivationf the ESPRI T-type algorithm which helps us understand how the ESPRIT algorithm works for depolarized components. Experimental results by using E-SAR data are also provided to show the performance of the ESPRIT estimation. Keywords-polarimetric SAR interferometry; forest canopy; ESPRIT algorithm; polarized and depolarized components In this paper, we derive an another derivation of the ESPRIT-based polarimetric SAR interferometry and show that how the depolarized component affects in the interferometric phase estimation. We derive theoretically that the ESPRIT estimation is biased for the data including depolarized components. However, we show by some numerical results that the bias due to depolarized components is sufficiently small when the depolarized component is smaller than the polarized components. At the L-band observations for forest, polarized ground components strong. This is why the ESPRIT algorithm is applicable to the L-band data.

Multi-baseline polarimetric SAR data classification using the complex Wishart distribution and principal component analysis

In this paper is introduced a classification approach for multi-baseline polarimetric interferometric SAR data sets, based on a principal component analysis of coherent scattering vectors. From the Wishart probability density function of a restricted data set, is defined a maximum likelihood decision rule to perform an iterative adaptive classification.

ALOS-PALSAR Quad. Pol. images and their applications

The four-component scattering power decomposition is applied to ALOS-PALSAR Quad. Pol. Data sets. PALSAR is Phased Array type L-band Synthetic Aperture Radar onboard ALOS (Advanced Land Observing Satellite) launched by JAXA, Japan, in 2006. It has acquired more than 10000 fully polarimetric scenes although the polarimetric mode is experimental one. This paper presents some of scenes in Sichuan earthquake area to see how polarimetric data works and to show the effectiveness and usefulness of fully polarimetric data.

Analysis of anisotropic behavior using sub-aperture polarimetric SAR data

In this paper, a fully polarimetric analysis method is introduced to decompose synthesized polarimetric images into sub-aperture data sets, which corresponds to the scene responses under different azimuthal look-angles. A statistical analysis of polarimetric parameters permits to clearly discriminate media showing a non-stationary behavior during the SAR integration. A method is proposed, which eliminates the influence of azimuthal backscattering variations in conventional polarimetric SAR data analysis.

An Introduction to Polarisation Effects in Wave Scattering and Their Applications in Target Classification

INTRODUCTION In this paper we consider the subject of scattering polarimetry. While the classical Jones calculus is now widely employed for the analysis of polarization transformation due to propagation through optical devices [Collet 1992], there is a relatively poor analytical appreciation of how to deal with the structure of depolarization effects caused by wave scattering. This despite a long history of depolarization measurements in astronomy [Dollfus 1992], remote sensing [Kouzoubov 1998], atmospheric optics [Mishchenko 2000] and surface roughness studies [Egan 1985]. This paper seeks to review progress in this field, to summarize the main aspects of our current analytical understanding of depolarization and to review important applications of these techniques in radar and optical remote sensing of targets. A key distinguishing feature of scattering polarimetry is the presence of depolarization, to be distinguished from cross-polarization. While the latter is a deterministic effect, which has been used for example in radar topographic mapping [Schuler 1996] and through its absence for buried mine detection [Carin 1998], the former is fundamentally a stochastic process whereby incident radiation is made noise-like due to the randomness of a scattering volume [Brosseaau 1998]. The classical methods for describing such phenomena are based on the degree of polarization, defined from the Stokes vector [Collet 1993]. However, there has been a recent trend towards using active sensors (particularly polarized Radar [Boerner 1992] and Lidar [Mishchenko 2000]) for remote sensing purposes. These systems can fully populate the scattering matrices of polarimetry and this has lead to a more sophisticated understanding of the generalised relationship between depolarization and scattering. This in turn has lead to improved methods for remote sensing using polarized waves [Boerner 1998]