Espace membre

Cet espace est dédié aux étudiants, aux enseignants et au personnel administratif de l'école


Mot de passe oublié?

Actualités de l'école


Doctorate thesis defense of Wajih Ben Abdallah

Doctorate thesis defense on March 31th 2016 at 9:00 AM ,in Amphi II, Sup’Com.

Entitled : Interferomteric phase unwrapping by adaptive modeling approach

Presented by : Wajih Ben Abdallah 



Mr. Mohamed Rached BOUSSEMA

Professor at ENIT, Tunisia





Mr. José Bioucas-Dias

Professor at Universidade de Lisboa, Portugal


Mr. Imed Riadh FARAH

Associate Professor at University of Manouba




Professor at SUPCOM




Professor at SUPCOM


The Interferometric Synthetic Aperture Radar (InSAR) is an interesting remote sensing technique to map the earth's surface and to follow its changes over large area. It is based on the interferometric phase difference (interferogram), computed between two complex SAR images, which is sensitive to the terrain topography.

However, due to many decorrelation factors, the interferogram is a noisy modulo 2π phase image. Then, it should be filtered and unwrapped in order to be exploitable and allows the generation of the Digital Elevation Model (DEM). Despite several works have been proposed in the literature to solve both problems, most of them are not fully taken into account the topographic and the geophysical properties of the studied areas to propose suitable solutions.

The present thesis comes within this context and it has three main specific objectives. First, we justify the choice between three possible strategies for SAR interferogram processing: a) the unwrapping without filtering, b) filtering before unwrapping and c) the simultaneous filtering and unwrapping. The choice of each strategy depends on the noise level in the observed interferogram and the topographic features of the acquired area. In the case of phase filtering, a decision rule to make choice between the spatial and the wavelet domains is proposed. The second objective is to propose filtering and unwrapping methods in the Markov Random Fields (MRF) framework. The parameters of the MRF models are adaptive to the acquisition features and the topographic and geophysical properties of each interferogram. The proposed approaches take benefit of the coherence map, generated from the SLC pair image that indicates the similarity degree between the two acquisitions. The third objective is to exploit the sparse recovery theory to propose filtering and unwrapping methods in the case of measurement missing due to the incoherent acquisitions. The estimation results of the sparse recovery based approaches are improved using the temporal correlation provided by the time series SAR interferograms as supplement information.


Interferometric SAR, phase unwrapping, phase filtering, coherence map, wavelet, Markov Random Field, sparse recovery erformance analysis.