To improve the evaluation of SAR image pairs, a special visualisation techniques has been developed. This method makes surface properties and changes accessible to an easy interpretation.
The so-called CovAmCohTM method represents a technique for the interactive analysis of complex valued (high resolution) spaceborne SAR image pairs. It permits the extraction of conclusions about surface types (physical backscatter information) and the detection of areas of temporal change or instable phase centers like volume scatterer e.g. represented by leafy vegetation.
CovAmCoh stands for the fusion of the three SAR image products coefficient of variation (Cov), intensity (Am2) and coherence (Coh) in one RGB image (R=Cov, G=Am, B=Coh). The coefficient of variation (ratio of standard deviation and mean value in a local pixel neighborhood) can be used as a measure for the local homogeneity, or the local heterogeneity (locality) in the intensity image. The intensity image contains information on the backscatter behavior of the earth's surface and objects placed on it. The spatial and temporal stability of the interferometric phase is represented by the coherence image.
With the help of this simple interpretation scheme in combination with the additive color layer mixing, typical scene areas could be detected in an interactive way. For example, temporarily instable areas could be detected by their lack of blue color due to the decorrelated interferometric phase. Very local heterogeneous high backscattering areas appear yellow (e.g. wastewater treatment plants, industrial tanks with swimming capture etc.). Otherwise, temporarily stable areas (coherent) are easy to detect by their blue color (e.g. railway lines with gravel between sleepers, flat roofs with gravel structure).
By comparing the CovAmCoh RGB image with the accordant amplitude image, it is obvious that there is a surplus profit of information which is available for the interpreter.