SAR image simulation for the assessment of despeckling techniques

Martino, Gerardo Di; Poderico, Mariana; Poggi, Gianfranco; Riccio, Daniele; Verdoliva, Luisa

doi:10.1109/igarss.2012.6351163

Cited by 5 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Complexity is measured by the algorithms' CPU time on a 3.40 GHz 64 bit desktop computer with 8 GB memory. To measure image quality we use three complementary approaches: statistical simulation based on optical images, the benchmarking tool proposed in [16], [17] based on physical-level simulation of SAR canonical scenes, and real-world SAR images.…”

Section: Resultsmentioning

confidence: 99%

Fast Adaptive Nonlocal SAR Despeckling

Cozzolino

Parrilli

Scarpa

et al. 2014

IEEE Geosci. Remote Sensing Lett.

Self Cite

167

101

View full text Add to dashboard Cite

Despeckling techniques based on the nonlocal approach provide an excellent performance, but exhibit also a remarkable complexity, unsuited to time-critical applications. In this letter, we propose a fast nonlocal despeckling filter. Starting from the recent SAR-BM3D algorithm, we propose to use a variable-size search area driven by the activity level of each patch, and a probabilistic early termination approach that exploits speckle statistics in order to speed up block matching. Finally, the use of look-up tables helps in further reducing the processing costs. The technique proposed conjugates excellent performance and low complexity, as demonstrated on both simulated and real-world SAR images and on a dedicated SAR despeckling benchmark.

show abstract

Section: Resultsmentioning

confidence: 99%

Fast Adaptive Nonlocal SAR Despeckling

Cozzolino

Parrilli

Scarpa

et al. 2014

IEEE Geosci. Remote Sensing Lett.

Self Cite

167

101

View full text Add to dashboard Cite

show abstract

“…A reliable statistical characterization of the speckle is of key importance for a huge set of applications, e.g. model-based despeckling [Di Martino et al, 2012a, 2013a and segmentation [Collins and Allan, 2009]. Therefore, a key parameter for the statistical characterization of the speckle is the number of independent scatterers per resolution cell, N. Under the hypothesis that N>>1, the central limit theorem can be applied giving rise to a Gaussian complex circular process, with an Exponential distributed intensity and a phase uniformly distributed in (0.2π).…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, it is crucial to have efficient tools able to predict the SAR data behaviour as a function of the scene parameters. As a matter of fact, the use of a SAR simulator can provide value-added information for SAR data interpretation and a support for SAR processing techniques (e.g., image despeckling [Di Martino et al, 2012a, 2013a, segmentation [Lee and Jurkevich, 1989;Collins and Allan, 2009], change detection , sea target (and extended target) detection [Watts et al, 1990;Tello et al, 2007]). In this paper a SAR simulator able to provide images presenting the appropriate speckle statistics is introduced.…”

Section: Introductionmentioning

confidence: 99%

A Physical Approach for SAR Speckle Simulation: First Results

Martino

Iodice

Riccio

et al. 2013

European Journal of Remote Sensing

Self Cite

View full text Add to dashboard Cite

In this paper a SAR simulator able to provide images presenting appropriate speckle statistics is introduced. The simulator is able to generate both Exponential (fully developed speckle) and K-distributed speckle statistics, according to surface and radar parameters. It is based on sound physical models for the evaluation of the number of equivalent scatterers per resolution cell. The proposed simulator requires as inputs the radar, orbital and surface parameters. The statistics relevant to each single equivalent scatterer are then effectively generated as a 2-D scale mixture of Gaussians and the global statistics of the return from each resolution cell are obtained as the coherent sum of all the scatterers contributions. The rationale of the proposed simulation framework is detailed and meaningful results regarding the analysis of the first simulated images are discussed.

show abstract

Classification-based nonlocal SAR despeckling

Gragnaniello

Poggi

Verdoliva

2012

2012 Tyrrhenian Workshop on Advances in Radar and Remote Sensing (TyWRRS)

View full text Add to dashboard Cite

Nonlocal techniques represent the current state of the art in SAR despeckling, providing a good compromise between speckle reduction and preservation of relevant image features. Nonetheless, they are not free from problems, going from the loss of image features to the introduction of their own brand of artifacts, due to the inability to deal equally well with all types of imaged scenes. A possible tool to improve performance is a prior segmentation or classification of the image, so as to adjust the filter parameters to fit the nature of the region under analysis. This work first provides some insight into the potential of classification-based nonlocal filtering by running simulation experiments in a controlled environment. Then proposes a new version of the SAR-BM3D despeckling technique in which each pixel is first classified as homogeneous or not, and then filtered with class-adapted parameters. Although results on real SAR images are still questionable, there is already some significant gain in selected areas that justifies the interest towards this approach

show abstract

SAR image simulation for the assessment of despeckling techniques

Cited by 5 publications

References 10 publications

Fast Adaptive Nonlocal SAR Despeckling

Fast Adaptive Nonlocal SAR Despeckling

A Physical Approach for SAR Speckle Simulation: First Results

Classification-based nonlocal SAR despeckling

Contact Info

Product

Resources

About