Search

Synthetic aperture radar (SAR) systems are efficient to deal with remote sensing issues. In contrast, SAR images are affected by speckle noise, due to the use of coherent illumination in their capturing. This noise imposes both a granular interference on such images (precluding their interpretability) and a multiplicative and non-Gaussian nature on their data. The multiplicative modelling is often used to surpass previous difficulties, mainly its particular case the G I 0 distribution. In this paper, we introduce a new time series modelling for SAR imagery, called G I 0 autoregressive-moving-average (ARMA) process. We derive some of its mathematical properties: score vector, Fisher information matrix, residual analysis and prediction equations. The maximum likelihood estimation procedure for G I 0 -ARMA parameters is discussed and some asymptotic behaviours for its estimates are quantified by Monte Carlo experiments. Applications to Munich and Foulum SAR actual images are made. Results show the proposed model can outperform the Γ-ARMA model. [ABSTRACT FROM AUTHOR]

The proposal of efficient classification methods is often required when common conditions (like additivity and normal stochastic behaviour) are not satisfied. Three classical classifiers are the Linear Discriminant Analysis (LDA), K Nearest Neighbours (KNN) and Quadratic Discriminant Analysis (QDA) methods. It is known that the performance of these techniques is strongly affected by the absence of linearity in the separation between/among two or more multivariate data classes. In this paper, we propose semiparametric classification methods that can be lesser sensible to the previous phenomenon. Our classifiers are based on LDA and KNN methods combined with Rényi and Kullback–Leibler stochastic divergences. The performance of various classifiers is evaluated on both normal and non-normal simulated data Further, they are applied to classify multidimensional features of synthetic aperture radar images, which have a multiplicative and non-normal nature due to the speckle noise presence. Results from empirical and real data furnish evidence that proposed methods can provide classification with smaller error rates than the classical LDA, KNN and QDA classification procedures. In particular, the optimum performance happens when order parameter tends to 0.5 for simulated data, while the best classification is achieved at order around 0.95 for real data. [ABSTRACT FROM AUTHOR]

We study a new family of continuous distributions with two extra shape parameters called the Burr generalized family of distributions. We investigate the shapes of the density and hazard rate function. We derive explicit expressions for some of its mathematical quantities. The estimation of the model parameters is performed by maximum likelihood. We prove the flexibility of the new family by means of applications to two real data sets. Furthermore, we propose a new extended regression model based on the logarithm of the Burr generalized distribution. This model can be very useful to the analysis of real data and provide more realistic fits than other special regression models. [ABSTRACT FROM PUBLISHER]