The paper considers the statistical modelling of fully developed backscattering in the case of SAR images of the ocean surface. According to the random-walk theory, the SAR image grey level is modelled as the product of a speckle noise and a variable which is dependent on the reflectivity of the illuminated surface and the radar-point-spread function. The purpose of the study is the statistical modelling of the latter variable. As nothing is known about these statistics, the authors propose the use of an estimation method based on a system of distributions. The set contains known density-probability functions with very flexible shapes that are supposed to fit its distribution. The associated image intensity distributions are processed and form a new system called KUBW, referring to the special functions used to generate the distributions. The classical K law belongs to the new system of distributions. By using a statistical test on the intensity distribution, the authors assess the relevance of the system of distributions in comparison with the classical model. The paper concludes with a discussion of the merits of the method and its extension to the case of ocean SAR image applications.
[1]
J. Goodman.
Statistical Properties of Laser Speckle Patterns
,
1963
.
[2]
S. F. George,et al.
Detection of Targets in Non-Gaussian Sea Clutter
,
1970,
IEEE Transactions on Aerospace and Electronic Systems.
[3]
P. R. Fisk,et al.
Distributions in Statistics: Continuous Multivariate Distributions
,
1971
.
[4]
E. Jakeman,et al.
A model for non-Rayleigh sea echo
,
1976
.
[5]
E. Jakeman,et al.
Significance of K Distributions in Scattering Experiments
,
1978
.
[6]
K. Ward.
Compound representation of high resolution sea clutter
,
1980
.
[7]
Yves Delignon,et al.
Unsupervised Bayesian segmentation of SAR images using the Pearson system distributions
,
1993,
Proceedings of IGARSS '93 - IEEE International Geoscience and Remote Sensing Symposium.
[8]
Abdelwaheb Marzouki,et al.
Estimation of generalized mixtures and its application in image segmentation
,
1997,
IEEE Trans. Image Process..