Camilla Brekke

Oil Spill Detection in Radarsat and Envisat SAR Images

We present algorithms for automatic detection of oil spills in synthetic aperture radar (SAR) images. The algorithms consist of three main parts, namely: 1) detection of dark spots; 2) feature extraction from the dark spot candidates; and 3) classification of dark spots as oil spills or look-alikes. The algorithms have been trained on a large number of Radarsat and Envisat Advanced Synthetic Aperture Radar (ASAR) images. The performance of the algorithm is compared to manual and semiautomatic approaches in a benchmark study using 59 Radarsat and Envisat images. The algorithms can be considered to be a good alternative to manual inspection when large ocean areas are to be inspected

Characterization of Marine Surface Slicks by Radarsat-2 Multipolarization Features

In this paper, we study surface slick characterization in polarimetric C-band synthetic aperture radar (SAR) data. The objective is to identify the most powerful multipolarization SAR descriptors for mineral oil spill versus biogenic slick discrimination. A systematic comparison of eight well-known multipolarization features is provided. The analysis is performed on data that we collected during a large-scale oil spill exercise at the Frigg field situated northwest of Stavanger, in June 2011. Controlled oil spills and simulated look-alikes were simultaneously captured within fine quad-polarization Radarsat-2 acquisitions during this experiment. Multipolarization features derived from only the copolarized complex scattering coefficients are explored. We find that the two most powerful multipolarization features extracted from this data set are the geometric intensity, measuring the combined intensity based on the determinant of the coherency matrix, and the real part of the copolarization cross product, which is related to the scattering behavior of the target. We show that these two features can distinguish between the simulated biogenic slicks and mineral oil types such as Balder and Oseberg blend, and that the discriminative power seems to be persistent with time.

Comparing Near-Coincident C- and X-Band SAR Acquisitions of Marine Oil Spills

In this paper, we compare satellite-borne Cand X-band synthetic aperture radar (SAR) data for marine oil spill observation. During large-scale oil-on-water exercises in the North Sea, quad-polarization Radarsat-2 (C-band) and dual-polarization TerraSAR-X (X-band) data were acquired with temporal distances of less than 24 min. The objective is to characterize and quantify differences in the Radarsat-2 and TerraSAR-X measurements. Three scene pairs are compared in terms of data quality and signal characteristics, including statistical properties and selected multipolarization (HH, VV) parameters. The signal characteristics are also compared among low-backscatter features of various origin within the individual pairs. No viable argument for selecting one sensor above the other is identified in the data quality study. In the statistical analysis, investigation of logcumulants indicates a larger deviation from Gaussian statistics in the TerraSAR-X data compared with Radarsat-2 measurements. Log-cumulant diagrams are also shown to be a useful tool for discrimination between oil spills and a simulated biogenic slick in both sensors. Multipolarization features show enhanced slick-sea contrasts and a better discrimination between mineral oil spills and other low-backscatter features in Radarsat-2 compared with TerraSAR-X. The presence of a non-Bragg scattering component in the data is revealed for both sensors. The relative contribution of non-Bragg scattering to the total backscatter is found to be higher in the TerraSAR-X data than in the Radarsat-2 data. In general, the non-Bragg component is found to account for a larger part of the backscatter in slick-covered areas compared with clean sea.

Classifiers and Confidence Estimation for Oil Spill Detection in ENVISAT ASAR Images

An improved classification approach is proposed for automatic oil spill detection in synthetic aperture radar images. The performance of statistical classifiers and support vector machines is compared. Regularized statistical classifiers prove to perform the best on this problem. To allow the user to tune the system with respect to the tradeoff between the number of true positive alarms and the number of false positives, an automatic confidence estimator has been developed. Combining the regularized classifier with confidence estimation leads to acceptable performance.

Robust CFAR Detector Based on Truncated Statistics in Multiple-Target Situations

A new and robust constant false alarm rate (CFAR) detector based on truncated statistics (TSs) is proposed for ship detection in single-look intensity and multilook intensity synthetic aperture radar data. The approach is aimed at high-target-density situations such as busy shipping lines and crowded harbors, where the background statistics are estimated from potentially contaminated sea clutter samples. The CFAR detector uses truncation to exclude possible statistically interfering outliers and TSs to model the remaining background samples. The derived truncated statistic CFAR (TS-CFAR) algorithm does not require prior knowledge of the interfering targets. The TS-CFAR detector provides accurate background clutter modeling, a stable false alarm regulation property, and improved detection performance in high-target-density situations.

Ship Detection in Ice-Infested Waters Based on Dual-Polarization SAR Imagery

This letter discusses the potential of automatic ship detection in ice-infested waters based on satellite synthetic aperture radar (SAR) imagery. The popular K -distribution is used to model the backscatter statistics of sea ice clutter. The goodness of fit of this model is assessed with the Kolmogorov-Smirnov and Anderson-Darling test statistics for both VV and VH polarizations. We also test the impact of introducing the Method of Log Cumulant (MoLC) estimator for the shape parameter of the K-distribution. Finally, a constant false-alarm rate ship detection algorithm, applying the K -distribution with the MoLC estimator, is evaluated on dual-polarization RADARSAT-2 SAR data. Our results demonstrate that this is a viable approach to ship detection in ice-infested waters.

Subband Extraction Strategies in Ship Detection With the Subaperture Cross-Correlation Magnitude

The subaperture cross-correlation magnitude (SCM) has previously been proposed as a statistic that improves the contrast between small ship targets and the surrounding sea in synthetic-aperture-radar images. This preprocessing technique utilizes the fast decorrelation of open-water surface ripples on the scale of the SAR wavelength relative to coherent targets such as a ship. However, optimization of the bandwidth splitting in the subband extraction has not received any attention. The aim of this letter is twofold: 1) to describe the technical details of the algorithm, including modifications that are necessary to allow overlapping subapertures; and 2) to study the effect of splitting the bandwidth into two azimuth subapertures with respect to varying bandwidths and subaperture overlap. The impact on the SCM is investigated in terms of measures of speckle reduction and target-to-clutter contrast. Experiments are performed on real single-look complex SAR data containing repeated acquisitions of a vessel in open sea. The results indicate that the subband extraction strategy has a clear impact on performance.

A Segmentation-Based CFAR Detection Algorithm Using Truncated Statistics

Target detection in nonhomogeneous sea clutter environments is a complex and challenging task due to the capture effect from interfering outliers and the clutter edge effect from background intensity transitions. For synthetic aperture radar (SAR) measurements, those issues are commonly caused by multiple targets and meteorological and oceanographic phenomena, respectively. This paper proposes a segmentation-based constant false-alarm rate (CFAR) detection algorithm using truncated statistics (TS) for multilooked intensity (MLI) SAR imagery, which simultaneously addresses both issues. From our previous work, TS is a useful tool when the region of interest (ROI) is contaminated by multiple nonclutter pixels. Within each ROI confined by the reference window, the proposed scheme implements an automatic image segmentation algorithm, which performs a finite mixture model estimation with a modified expectation-maximization algorithm. Data truncation is applied here to exclude all possible statistically interfering classes, and sample modeling is based upon the truncated two-parameter gamma model. Next, CFAR detection is conducted pixel by pixel, utilizing the statistical information obtained from the segmentation process within the local reference window. The segmentation-based CFAR detection scheme is examined with real Radarsat-2 MLI SAR imagery. Compared with the conventional CFAR detection approaches, our proposal provides improved background clutter modeling and robust detection performance in nonhomogeneous clutter environments.

Feature extraction for oil spill detection based on SAR images

Algorithms based on SAR images for the purpose of detecting illegal oil spill pollution in the marine environment are studied. This paper focus on the feature extraction step, aiming at identifying features that lead to significant improvements in classification performance compared to earlier reported results. Both traditional region descriptors, features tailored to oil spill detection and techniques originally associated with other applications are evaluated. Experimental results show an increase from 89% to 97% in the number of suspected oil spills detected.

Measurement and modeling of oil slick transport

Transport characteristics of oil slicks are reported from a controlled release experiment conducted in the North Sea in June 2015, during which mineral oil emulsions of different volumetric oil fractions and a look-alike biogenic oil were released and allowed to develop naturally. The experiment used the Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) to track slick location, size, and shape for ∼8 hours following release. Wind conditions during the exercise were at the high end of the range considered suitable for radar-based slick detection, but the slicks were easily detectable in all images acquired by the low noise, L-band imaging radar. The measurements are used to constrain the entrainment length and representative droplet radii for oil elements in simulations generated using the OpenOil advanced oil drift model. Simultaneously released drifters provide near-surface current estimates for the single biogenic release and one emulsion release, and are used to test model sensitivity to upper ocean currents and mixing. Results of the modeling reveal a distinct difference between the transport of the biogenic oil and the mineral oil emulsion, in particular in the vertical direction, with faster and deeper entrainment of significantly smaller droplets of the biogenic oil. The difference in depth profiles for the two types of oils is substantial, with most of the biogenic oil residing below depths of 10 m, compared to the majority of the emulsion remaining above 10 m depth. This difference was key to fitting the observed evolution of the two different types of slicks. This article is protected by copyright. All rights reserved.