پديدآورندگان :
Zakeri S. zakeri.salar@ut.ac.ir School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran, Iran , Motagh M. motagh@gfz.post 1 School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran, Iran
2 GFZ German Research Center for Geosciences, Department of Geodesy, Section of Remote Sensing, Potsdam, Germany , Merryman Boncori J.P. Sarmap S.A , Pasquali P. Sarmap S.A , Safari A. aasafari@ut.ac.ir School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran, Iran
كليدواژه :
SAR tomography , persistent scatterer , synthetic aperture radar , compressive sensing , Urban , layover , Non , linear least square
چكيده فارسي :
SAR sensors have a side-looking imaging system and due to this side-looking geometry in urban areas the acquired images are affected with lay-over problem. To overcome this layover problem and resolve the scatteres located in the same pixel several tomography methods have been developed. In this paper we compare the capabilities and advantages of three different methods of tomographic processing both on simulated data and real data. The methods include first order model, Nonlinear least-square (NLS) and SL1MMER (Scale-down by L1 norm Minimization, Model selection, and Estimation Reconstruction). Results show that NLS could provide a good accuracy. However one needs to have a prior knowledge about the model order. On the other hand the SL1MMER method over comes this obstacle using the sparsity of the reflectivity along the height combined with a cost function providing accurate and fast solution.
