Extraction of textural information in polarimetric SAR

The main purpose of this paper has been to show the potential of using the covariance matrix eigenvalues in order to estimate the texture information of span polarimetric images. Comparison with experimental results from the JPL airborne SAR campaign demonstrates the capability of the eigenvalues of predicting the scattering behaviour of the scene through a texture analysis. In the case of low signal-to-noise ratio, the texture variance of the span polarimetric image deteriorates strongly, suggesting the use of the covariance matrix information to predict sensor anomalies. Future research will be devoted to the analysis of non-Gaussian clutter and to a quantitative formulation of the effects of sensor parameters (signal-to-noise, crosstalk, etc.) on the covariance matrix eigenvalues