Statistical Analysis of High-Resolution SAR Ground Clutter Data (original) (raw)

A Noncentral and Non-Gaussian Probability Model for SAR Data

Image Analysis, 2017

A general compound statistical model for coherent imaging is developed and tested on single-channel Synthetic Aperture Radar (SAR) data. In this formulation, coherent scattering is taken into consideration and the texture is modeled using an Inverse Gaussian distribution. Parameter estimation is conducted via an Expectation Maximization (EM) scheme. A Maximum a Posteriori (MAP) speckle filter based on this model is also implemented. The filter shows good smoothing capabilities and preserves details in the selected scene, showing promise for targetdetection applications.

The effect of speckle filtering on SAR texture discrimination

Simpósio Brasileiro de …, 1996

In tropical ecology studies, forest classification is a key issue.Although there is no widely accepted forest classification criterion, it is recognized that texture is an important factor to discriminate forest types and other land cover, particularly when using radar images. Synthetic Aperture Radar (SARA) images, however, are contaminated by a multiplicative noise, known as speckle, wich disturbs the texture identification. Several filters have been proposed to atenuate this kind of noise, but the effect of these filters on texture is not well known. In this paper, textures are modelled by two-dimensional autorregressive (AR-2D) models. These models are estimated for each one of the samples of SAR textures before and after speckle filtering. Eight samples of primary forest and seven samples of non-forest (pasture and agricultural crops) were collect from SAREX data (Cband, HH polarization, 6m resolution, 6 looks) in the Tapajós National Forest (Flona) region in Pará state, Brazil. All these samples were filtered by a 3 x 3 Box filter and a 3 x 3 Frost filter. Euclidean distances were computed between the model coefficient vector of the samples and the average coefficient vector for the two classes (defined here as the class vectors) for the unfiltered and filtered cases separately. For all cases the coefficient vectors formed two separated clusters, corresponding to each one of the classes, in a non-linear mapping of the coefficient space. The conclusions are: AR modelling is an effective method to idendify and discriminate radar texture. The discriminatory power, however, is higher using the unfiltered channels than when the simple Box and the Frost filter are used.

A model for removal of speckle noise in SAR images (ALOS PALSAR)

Canadian Journal of Remote Sensing, 2008

Speckle noise is primarily due to the phase fluctuations of the electromagnetic return signals. Since inherent spatial-correlation characteristics of speckle in synthetic aperture radar (SAR) images are not exploited in existing multiplicative models for speckle noise, a speckle noise model is proposed here that provides a new framework for modelling and reducing the speckle noise. Both quantitative and qualitative criteria, including speckle reduction and texture preservation, are used to evaluate the performance of the proposed filter; one PALSAR (new Japanese sensor) image and a JERS-1 image are employed in the evaluation. The results showed that the proposed filter is slightly better than commonly used filters such as the Kuan, gamma, enhanced Lee, and enhanced Frost filters. The proposed filter can be used in different applications, including mapping and forestry biomass estimation. Furthermore, one of the benefits of the proposed filter is that it is independent of the threshold, which is required in most commonly used filters. The proposed filter was tested with SAR images of different sites in the northern forests of Iran. Résumé. Le chatoiement est un bruit qui est principalement dû aux fluctuations de phase des signaux électromagnétiques de retour. Comme les caractéristiques inhérentes de corrélation spatiale du chatoiement des images RSO ne sont pas exploitées dans les modèles multiplicatifs actuels de chatoiement, on propose dans cet article un modèle de chatoiement qui fournit un nouveau cadre pour la modélisation et la réduction du chatoiement. Des critères à la fois quantitatifs et qualitatifs, incluant la réduction du chatoiement et la préservation de la texture sont utilisés pour évaluer la performance du filtre proposé. Une image de PALSAR (nouveau capteur japonais) ainsi qu'une image de JERS-1 sont utilisées pour réaliser l'évaluation. Les résultats ont montré que le filtre proposé est légèrement meilleur que les filtres utilisés couramment tels que les filtres de Kuan, Gamma, de Lee amélioré et de Frost amélioré. Le filtre proposé peut être utilisé dans différentes applications, incluant la cartographie et la foresterie dans le contexte de l'estimation de la biomasse. De plus, un des avantages du filtre proposé est qu'il est indépendant du seuil qui est normalement requis dans le cas des filtres les plus couramment utilisés. Le filtre proposé a été testé sur différents sites d'images RSO des zones forestières du nord de l'Iran. [Traduit par la Rédaction] Sumantyo and Amini 515

Classification of SAR images using a general and tractable multiplicative model

International Journal of Remote Sensing, 2003

Among the frameworks for Synthetic Aperture Radar (SAR) image modelling and analysis, the multiplicative model is very accurate and successful. It is based on the assumption that the observed random field is the result of the product of two independent and unobserved random fields: X and Y . The random field X models the terrain backscatter and, thus, depends only on the type of area each pixel belongs to. The random field Y takes into account that SAR images are the result of a coherent imaging system that produces the well known phenomenon called speckle noise, and that they are generated by performing an average of n statistically independent images (looks) in order to reduce the noise effect. There are various ways of modelling the random field X; recently Frery, Müller, Yanasse & Sant'Anna (1997) proposed the Γ −1/2 (α, γ) distribution. This, with the usual Γ 1/2 (n, n) distribution for the amplitude speckle, resulted in a new distribution for the return: the G 0 A (α, γ, n) law. The parameters α and γ depend only on the ground truth, and n is the number of looks. The advantage of this distribution over the ones used in the past is that it models very well extremely heterogeneous areas like cities, as well as moderately heterogeneous areas like forests, and homogeneous areas like pastures as well. As the ground truth can be characterized by the parameters α and γ, their estimation in each pixel generates parameter maps that can be used as the input for classification methods. In this work, moment estimators are used on simulated and on real SAR images and, then, a supervised classification technique (Gaussian maximum likelihood) is performed and evaluated. Excellent classification results are obtained.

Modeling non-Rayleigh speckle distribution in SAR images

IEEE Transactions on Geoscience and Remote Sensing, 2002

In non-Rayleigh distributed radar images, the number of scatterers can be viewed as a Poisson distributed random variable, with the mean itself random. When this mean is Gamma distributed, then the image classically satisfies the distribution. We add three new possible distributions for this mean: inverse Gamma, Beta of the first kind, and Beta of the second kind. We show that new intensity distributions so obtained can be estimated, with the interest of the extension validated on a real image.

Markovian classification of SAR images using mathcalG0I\mathcal{G}^{0}_{I}mathcalG0I model

Brazilian Journal of Probability and Statistics, 2009

When processing synthetic aperture radar (SAR) images, there is a strong need for statistical models of scattering to take into account multiplicative noise. For instance, the classification process needs to be based on the use of statistics. Our main contribution is the choice of an accurate model for SAR images over urban areas and its use in a Markovian classification algorithm. Clutter in SAR images becomes non-Gaussian when the resolution is high or when the area is manmade. Many models have been proposed to fit with non-Gaussian scattering statistics (K, Weibull, Log-normal, etc.), but none of them is flexible enough to model all kinds of surfaces. Frery et al. [IEEE Transactions on Geoscience and Remote Sensing 35 (1997) 648-659] proposed a new class of distributions, G distribution, arising from the multiplicative model. Classical distributions such as K are particular cases of this new class. A special case of this class called G 0 is shown able to model extremely heterogeneous clutter, such as that of urban areas. The quality of the classification obtained by mixing this model and a Markovian segmentation is high.

Matched subspace detectors for discrimination of targets from trees in SAR imagery

2000

Several approaches that exploit forest clutter structure We investigate the use of subspace-based detectors for discriminating vehicles from trees in low frequency synthetic aperture imagery. We model tree scattering as structured isotropic interference responses and model dominant vehicle scattering as dihedral responses. We form linear subspaces of tree and target responses, and apply subspace-based detection methods developed by Scharf and Friedlander. Analysis on synthetic tree and target models show the viability of this approach. Preliminary results on measured imagery provide lower performance, suggesting the need for improved data calibration and improved scattering models of trees at low frequencies.

A Statistical Analysis for Wavelength-Resolution SAR Image Stacks

IEEE Geoscience and Remote Sensing Letters, 2019

This letter presents a clutter statistical analysis for stacks of wavelength-resolution synthetic aperture radar (SAR) images. Each image stack consists of SAR images generated by the same sensor, using the same flight track illuminating the same scene but with a time separation between the illuminations. We test three candidate statistical distributions for time changes in the stack, namely, Rician, Rayleigh, and log-normal. The tests results reveal that the Rician distribution is a very good candidate for modeling stack of wavelength-resolution SAR images, where 98.59% of the tested samples passed the Anderson-Darling (AD) goodness-of-fit test. Also, it is observed that the presence of changes in the ground scene is related to the tested samples that have failed in the AD test for the Rician distribution hypothesis. Index Terms-Anderson-Darling (AD) test, CARABAS II, change detection, image stack, multitemporal synthetic aperture radar (SAR) images, SAR. I. INTRODUCTION I T is well known that synthetic aperture radar (SAR) systems are used to monitor global climate changes, to observe and predict natural disasters, and for several other applications based on observations of the earth. SAR sensors can be mounted on both orbital platforms or airborne. The latter is desirable, for example, when shorter observation Manuscript