InSAR coherence estimation for temporal analysis and phase unwrapping applications

In SAR interferometry, coherence between data acquisitions is potential information on the accuracy of interferograms. In this paper, a new coherence estimation approach is investigated. Our study is based on a binary mixture modelisation of coherence images. A Bayesian criteria is so used to discriminate between the two statistical mixed laws. It is shown that empirical coherence is biased and depends on the number of pixels averaged. The results of this paper can be applied to the estimation of the interferogram noise in interferometric SAR.