-
Regression Model for Speckled Data with Extremely Variability
Authors:
A. D. C. Nascimento,
J. M. Vasconcelos,
R. J. Cintra,
A. C. Frery
Abstract:
Synthetic aperture radar (SAR) is an efficient and widely used remote sensing tool. However, data extracted from SAR images are contaminated with speckle, which precludes the application of techniques based on the assumption of additive and normally distributed noise. One of the most successful approaches to describing such data is the multiplicative model, where intensities can follow a variety o…
▽ More
Synthetic aperture radar (SAR) is an efficient and widely used remote sensing tool. However, data extracted from SAR images are contaminated with speckle, which precludes the application of techniques based on the assumption of additive and normally distributed noise. One of the most successful approaches to describing such data is the multiplicative model, where intensities can follow a variety of distributions with positive support. The $\mathcal{G}^0_I$ model is among the most successful ones. Although several estimation methods for the $\mathcal{G}^0_I$ parameters have been proposed, there is no work exploring a regression structure for this model. Such a structure could allow us to infer unobserved values from available ones. In this work, we propose a $\mathcal{G}^0_I$ regression model and use it to describe the influence of intensities from other polarimetric channels. We derive some theoretical properties for the new model: Fisher information matrix, residual measures, and influential tools. Maximum likelihood point and interval estimation methods are proposed and evaluated by Monte Carlo experiments. Results from simulated and actual data show that the new model can be helpful for SAR image analysis.
△ Less
Submitted 14 October, 2024;
originally announced October 2024.
-
The Gamma Generalized Normal Distribution: A Descriptor of SAR Imagery
Authors:
G. M. Cordeiro,
R. J. Cintra,
L. C. Rêgo,
A. D. C. Nascimento
Abstract:
We propose a new four-parameter distribution for modeling synthetic aperture radar (SAR) imagery named the gamma generalized normal (GGN) by combining the gamma and generalized normal distributions. A mathematical characterization of the new distribution is provided by identifying the limit behavior and by calculating the density and moment expansions. The GGN model performance is evaluated on bot…
▽ More
We propose a new four-parameter distribution for modeling synthetic aperture radar (SAR) imagery named the gamma generalized normal (GGN) by combining the gamma and generalized normal distributions. A mathematical characterization of the new distribution is provided by identifying the limit behavior and by calculating the density and moment expansions. The GGN model performance is evaluated on both synthetic and actual data and, for that, maximum likelihood estimation and random number generation are discussed. The proposed distribution is compared with the beta generalized normal distribution (BGN), which has already shown to appropriately represent SAR imagery. The performance of these two distributions are measured by means of statistics which provide evidence that the GGN can outperform the BGN distribution in some contexts.
△ Less
Submitted 3 June, 2022;
originally announced June 2022.
-
A new extended Cardioid model: an application to wind data
Authors:
Fernanda V. Paula,
Abraão D. C. Nascimento,
Getúlio J. A. Amaral
Abstract:
The Cardioid distribution is a relevant model for circular data. However, this model is not suitable for scenarios were there is asymmetry or multimodality. In order to overcome this gap, an extended Cardioid model is proposed, which is called Exponentiated Cardioid (EC) distribution. Besides, some of its properties are derived, such as trigonometric moments, kurtosis and skewness. A discussion ab…
▽ More
The Cardioid distribution is a relevant model for circular data. However, this model is not suitable for scenarios were there is asymmetry or multimodality. In order to overcome this gap, an extended Cardioid model is proposed, which is called Exponentiated Cardioid (EC) distribution. Besides, some of its properties are derived, such as trigonometric moments, kurtosis and skewness. A discussion about the modality and and expressions for the quantiles through approximations of the studied model are also presented. To fit the EC model, two estimation methods are presented based on maximum likelihood and quantile least squares procedures. The performance of proposed estimators is evaluated in a Monte Carlo simulation study, adopting both bias and mean square error as comparison criteria. Finally, the proposed model is applied to a dataset in the wind direction context. Results indicate that the EC distribution may outperform Cardioid and the von Mises distributions.
△ Less
Submitted 5 December, 2017;
originally announced December 2017.
-
Bias Correction and Modified Profile Likelihood under the Wishart Complex Distribution
Authors:
Abraão D. C. Nascimento,
Alejandro C. Frery,
Renato J. Cintra
Abstract:
This paper proposes improved methods for the maximum likelihood (ML) estimation of the equivalent number of looks $L$. This parameter has a meaningful interpretation in the context of polarimetric synthetic aperture radar (PolSAR) images. Due to the presence of coherent illumination in their processing, PolSAR systems generate images which present a granular noise called speckle. As a potential so…
▽ More
This paper proposes improved methods for the maximum likelihood (ML) estimation of the equivalent number of looks $L$. This parameter has a meaningful interpretation in the context of polarimetric synthetic aperture radar (PolSAR) images. Due to the presence of coherent illumination in their processing, PolSAR systems generate images which present a granular noise called speckle. As a potential solution for reducing such interference, the parameter $L$ controls the signal-noise ratio. Thus, the proposal of efficient estimation methodologies for $L$ has been sought. To that end, we consider firstly that a PolSAR image is well described by the scaled complex Wishart distribution. In recent years, Anfinsen et al. derived and analyzed estimation methods based on the ML and on trace statistical moments for obtaining the parameter $L$ of the unscaled version of such probability law. This paper generalizes that approach. We present the second-order bias expression proposed by Cox and Snell for the ML estimator of this parameter. Moreover, the formula of the profile likelihood modified by Barndorff-Nielsen in terms of $L$ is discussed. Such derivations yield two new ML estimators for the parameter $L$, which are compared to the estimators proposed by Anfinsen et al. The performance of these estimators is assessed by means of Monte Carlo experiments, adopting three statistical measures as comparison criterion: the mean square error, the bias, and the coefficient of variation. Equivalently to the simulation study, an application to actual PolSAR data concludes that the proposed estimators outperform all the others in homogeneous scenarios.
△ Less
Submitted 18 April, 2014;
originally announced April 2014.
-
Information Theory and Image Understanding: An Application to Polarimetric SAR Imagery
Authors:
A. C. Frery,
A. D. C. Nascimento,
R. J. Cintra
Abstract:
This work presents a comprehensive examination of the use of information theory for understanding Polarimetric Synthetic Aperture Radar (PolSAR) images by means of contrast measures that can be used as test statistics. Due to the phenomenon called `speckle', common to all images obtained with coherent illumination such as PolSAR imagery, accurate modelling is required in their processing and analy…
▽ More
This work presents a comprehensive examination of the use of information theory for understanding Polarimetric Synthetic Aperture Radar (PolSAR) images by means of contrast measures that can be used as test statistics. Due to the phenomenon called `speckle', common to all images obtained with coherent illumination such as PolSAR imagery, accurate modelling is required in their processing and analysis. The scaled multilook complex Wishart distribution has proven to be a successful approach for modelling radar backscatter from forest and pasture areas. Classification, segmentation, and image analysis techniques which depend on this model have been devised, and many of them employ some kind of dissimilarity measure. Specifically, we introduce statistical tests for analyzing contrast in such images. These tests are based on the chi-square, Kullback-Leibler, Rényi, Bhattacharyya, and Hellinger distances. Results obtained by Monte Carlo experiments reveal the Kullback-Leibler distance as the best one with respect to the empirical test sizes under several situations which include pure and contaminated data. The proposed methodology was applied to actual data, obtained by an E-SAR sensor over surroundings of We$β$ssling, Bavaria, Germany.
△ Less
Submitted 8 February, 2014;
originally announced February 2014.
-
Analytic Expressions for Stochastic Distances Between Relaxed Complex Wishart Distributions
Authors:
Alejandro C. Frery,
Abraão D. C. Nascimento,
Renato J. Cintra
Abstract:
The scaled complex Wishart distribution is a widely used model for multilook full polarimetric SAR data whose adequacy has been attested in the literature. Classification, segmentation, and image analysis techniques which depend on this model have been devised, and many of them employ some type of dissimilarity measure. In this paper we derive analytic expressions for four stochastic distances bet…
▽ More
The scaled complex Wishart distribution is a widely used model for multilook full polarimetric SAR data whose adequacy has been attested in the literature. Classification, segmentation, and image analysis techniques which depend on this model have been devised, and many of them employ some type of dissimilarity measure. In this paper we derive analytic expressions for four stochastic distances between relaxed scaled complex Wishart distributions in their most general form and in important particular cases. Using these distances, inequalities are obtained which lead to new ways of deriving the Bartlett and revised Wishart distances. The expressiveness of the four analytic distances is assessed with respect to the variation of parameters. Such distances are then used for deriving new tests statistics, which are proved to have asymptotic chi-square distribution. Adopting the test size as a comparison criterion, a sensitivity study is performed by means of Monte Carlo experiments suggesting that the Bhattacharyya statistic outperforms all the others. The power of the tests is also assessed. Applications to actual data illustrate the discrimination and homogeneity identification capabilities of these distances.
△ Less
Submitted 19 April, 2013;
originally announced April 2013.
-
Entropy-based Statistical Analysis of PolSAR Data
Authors:
Alejandro C. Frery,
Renato J. Cintra,
Abraão D. C. Nascimento
Abstract:
Images obtained from coherent illumination processes are contaminated with speckle noise, with polarimetric synthetic aperture radar (PolSAR) imagery as a prominent example. With an adequacy widely attested in the literature, the scaled complex Wishart distribution is an acceptable model for PolSAR data. In this perspective, we derive analytic expressions for the Shannon, Rényi, and restricted Tsa…
▽ More
Images obtained from coherent illumination processes are contaminated with speckle noise, with polarimetric synthetic aperture radar (PolSAR) imagery as a prominent example. With an adequacy widely attested in the literature, the scaled complex Wishart distribution is an acceptable model for PolSAR data. In this perspective, we derive analytic expressions for the Shannon, Rényi, and restricted Tsallis entropies under this model. Relationships between the derived measures and the parameters of the scaled Wishart law (i.e., the equivalent number of looks and the covariance matrix) are discussed. In addition, we obtain the asymptotic variances of the Shannon and Rényi entropies when replacing distribution parameters by maximum likelihood estimators. As a consequence, confidence intervals based on these two entropies are also derived and proposed as new ways of capturing contrast. New hypothesis tests are additionally proposed using these results, and their performance is assessed using simulated and real data. In general terms, the test based on the Shannon entropy outperforms those based on Rényi's.
△ Less
Submitted 15 October, 2012;
originally announced October 2012.
-
Hypothesis Testing in Speckled Data with Stochastic Distances
Authors:
Abraão D. C. Nascimento,
Renato J. Cintra,
Alejandro C. Frery
Abstract:
Images obtained with coherent illumination, as is the case of sonar, ultrasound-B, laser and Synthetic Aperture Radar -- SAR, are affected by speckle noise which reduces the ability to extract information from the data. Specialized techniques are required to deal with such imagery, which has been modeled by the G0 distribution and under which regions with different degrees of roughness and mean br…
▽ More
Images obtained with coherent illumination, as is the case of sonar, ultrasound-B, laser and Synthetic Aperture Radar -- SAR, are affected by speckle noise which reduces the ability to extract information from the data. Specialized techniques are required to deal with such imagery, which has been modeled by the G0 distribution and under which regions with different degrees of roughness and mean brightness can be characterized by two parameters; a third parameter, the number of looks, is related to the overall signal-to-noise ratio. Assessing distances between samples is an important step in image analysis; they provide grounds of the separability and, therefore, of the performance of classification procedures. This work derives and compares eight stochastic distances and assesses the performance of hypothesis tests that employ them and maximum likelihood estimation. We conclude that tests based on the triangular distance have the closest empirical size to the theoretical one, while those based on the arithmetic-geometric distances have the best power. Since the power of tests based on the triangular distance is close to optimum, we conclude that the safest choice is using this distance for hypothesis testing, even when compared with classical distances as Kullback-Leibler and Bhattacharyya.
△ Less
Submitted 12 July, 2012;
originally announced July 2012.
-
Parametric and Nonparametric Tests for Speckled Imagery
Authors:
Renato J. Cintra,
Abraão D. C. Nascimento,
Alejandro C. Frery
Abstract:
Synthetic aperture radar (SAR) has a pivotal role as a remote imaging method. Obtained by means of coherent illumination, SAR images are contaminated with speckle noise. The statistical modeling of such contamination is well described according with the multiplicative model and its implied G0 distribution. The understanding of SAR imagery and scene element identification is an important objective…
▽ More
Synthetic aperture radar (SAR) has a pivotal role as a remote imaging method. Obtained by means of coherent illumination, SAR images are contaminated with speckle noise. The statistical modeling of such contamination is well described according with the multiplicative model and its implied G0 distribution. The understanding of SAR imagery and scene element identification is an important objective in the field. In particular, reliable image contrast tools are sought. Aiming the proposition of new tools for evaluating SAR image contrast, we investigated new methods based on stochastic divergence. We propose several divergence measures specifically tailored for G0 distributed data. We also introduce a nonparametric approach based on the Kolmogorov-Smirnov distance for G0 data. We devised and assessed tests based on such measures, and their performances were quantified according to their test sizes and powers. Using Monte Carlo simulation, we present a robustness analysis of test statistics and of maximum likelihood estimators for several degrees of innovative contamination. It was identified that the proposed tests based on triangular and arithmetic-geometric measures outperformed the Kolmogorov-Smirnov methodology.
△ Less
Submitted 10 July, 2012;
originally announced July 2012.
