M. Bertacca

A FARIMA-based technique for oil slick and low-wind areas discrimination in sea SAR imagery

This paper introduces a new analysis technique, using the fractionally integrated autoregressive-moving average (FARIMA) model, to distinguish between low-wind and oil slick areas in high-resolution sea synthetic aperture radar (SAR) imagery. The method deals with the estimation of the fractional differencing and autoregressive-moving average parameters of the mean radial power spectral density of sea SAR images. The algorithm is applied and validated on dark areas corresponding to oil slicks, oil spills, and low-wind sea surface anomalies in European Remote Sensing 1 and 2 Precision Images of the Mediterranean Sea, North Sea, and Atlantic Ocean.

Two-Dimensional Variation Algorithm for Fractal Analysis of Sea SAR Images

This paper proposes a novel algorithm for estimating the fractal dimension of sea synthetic aperture radar (SAR) images. The algorithm is based on the variation method, and it is suitably designed for the analysis of sea SAR images. The SAR image fractal dimension is a feature that provides a measure of the image roughness. Such a feature can play an important role in the classification process for recognizing the presence of anomalies on the sea surface. The innovation aspects of this paper are listed as follows: (1) an extension of the variation method, which was proposed for the fractal analysis of one-dimensional signals, to the case of two-dimensional (2-D) functions; (2) a numerical formulation of the variation method, which is suitable for processing 2-D discrete signals; and (3) an optimization of the algorithm for sea SAR image analysis. The algorithm is tested and validated both on simulated and real ERS-1/2 Precision Image sea SAR images and compared with the classical estimation algorithm based on spectral analysis

A FARIMA based analysis for wind falls and oil slicks discrimination in sea SAR imagery

The aim of this paper is to define an analysis technique, which uses the fractionally integrated autoregressive-moving average (FARIMA) model, for discriminating wind falls from oil slick areas in sea SAR imagery. The method deals with the estimation of the fractional differencing and ARMA parameters for the sea SAR images mean radial PSD and is applied to some ERS-1 and ERS-2 images of the Mediterranean sea, North sea and Atlantic (Galicia) ocean containing only oil slick or only wind falls, or wind falls and oil slick anomalies

A FEXP model Short Range Dependence analysis for improving oil slicks and low-wind areas discrimination in sea SAR imagery

Starting from a consideration of the Long Range Dependence (LRD) behavior of sea SAR image spectra, an overview is given of the LRD approaches currently being used to achieve reliable sea surface anomalies discrimination from high resolution sea SAR images. In this paper, the problem of SAR image analysis to discriminate oil slicks from low wind areas on the sea surface is addressed by employing fractional exponential (FEXP) models and short range dependence (SRD) parameters. The presented method demonstrated reliable results when applied to European remote sensing 2 (ERS-2) SAR precision images (PRI) and ERS-2 SAR ellipsoid geocoded images (GEC) of the Atlantic and the Pacific Oceans.

Homomorphic Inverse-FEXP filtering for target detection in sea SAR images

Generalized linear filtering techniques allow signals that have been nonadditively combined to be separated, and the analytical convenience of the principle of superposition for linear systems to be preserved. In this paper we employ Fractionally Exponential spectral models (FEXP) to adaptively modify homomorphic filtering of sea Synthetic Aperture Radar (SAR) images, thereby improving target detection in sea SAR imagery. The presented method demonstrated reliable results when applied to ERS1 and ERS2 SAR PRI of the Mediterranean Sea and North Sea.

Isotropic and Anisotropic FEXP-Fractal Spectral Models for High Resolution Sea SAR Images

The aim of this paper is the definition of an anisotropic self-similar spectral model for high-resolution sea synthetic aperture radar (SAR) imagery. The assumption of spatial isotropy for sea SAR images is valid when the sea is calm, as the sea wave energy is spread in all directions and the SAR image PSD shows a circular symmetry. However, if the surface wind speed exceeds 7 m/s the anisotropy of sea SAR images starts to be perceptible. In this work we define an anisotropic spectral model by adding a 2-D isotropic fractionally exponential model (FEXP) to an anisotropic term defined starting from the fractal model of sea surface spectra. For this FEXP-Fractal (FEXPF) model we define a new class of spreading functions to characterize the anisotropic component. FEXPF models allow the spectra of sea SAR images to be simulated in different sea states and wind conditions - and with oil slicks -at low computational costs. FEXPF demonstrated reliable results when applied to European Remote Sensing (ERS) SAR Precision Images (PRI) of the Mediterranean Sea and the AtlanticOcean.

Sea SAR Image Simulation using Isotropic and Anisotropic FEXP-Fractal Spectral Models

Starting from a consideration of the anisotropic structure of sea synthetic aperture radar (SAR) image spectra, an overview is given of the approaches currently being used to achieve reliable sea SAR image simulation using fractionally exponential and fractal spectral models (FEXPF). The presented method demonstrated reliable results when applied to European Remote Sensing 2 (ERS-2) SAR Precision Images (PRI) and ERS-2 SAR Ellipsoid Geocoded Images (GEC) of the Mediterranean Sea and the Atlantic Ocean.

Different homogeneity detectors for improving Space-Time Adaptive Radar performance in heterogeneous clutter

Secondary data selection for estimation of the clutter covariance matrix in space-time adaptive processing (STAP) is normally obtained from cells (range rings) in close proximity of the cell under test. The aim of this paper is the analysis of performance improvement of space-time adaptive radars when secondary data selection is obtained by discriminating between quasi-homogeneous areas on the ground which generate clutter with different statistics (i.e. clutter edges including littoral, farmland-wooded hills or rural-urban interfaces). The algorithm presented in this paper, referred to as the different homogeneity detector (DHD), has been tested with simulated data obtained by using a general clutter model and a uniform linear array.