The two SAR images were preprocessed to create single-look complex images (SLC), including phase and amplitude information. Then, the SLC images were co-registered between the master and slave images (images before and after the earthquake) into the same geometry, and the master image was multiplied by the complex conjugate of the ‘slave’ image to form the interferogram. The signal from topography was removed by calculating a digital elevation model from NASA's Shuttle Radar Topography Mission (SRTM data). The interferogram was filtered using the Goldstein adaptive filtering method to focus on the most important signals. The minimum cost flow method was used to convert the signal's phase component from modulo 2π values to a continuous function, resulting in the total phase shift between the ground and the satellite. Following that, this unwrapped interferogram was used to refine the viewing geometry and thus refine the orbital topographic corrections. Finally, the interferogram was geocoded (WGS84) to produce the image provided by (Chen and Zebker, 2000; Wegnüller et al., 2016).
The MAI approach introduces forward- and backward-looking residual interferograms resulting from the phase difference between a full-aperture differential interferogram and a sub-aperture differential interferogram (forward- and backward-looking differential interferograms). This method reduces phase noise in incoherent areas by combining two residual interferograms. The ALOS-2 (L-band) and Sentinel-1 (C-band) data were collected in descending orbit. The ENVI seascape software v5.3 was used to process data
Processing
The six main steps of the MAI method processing are listed below and are described in (Figure 2.8)
Generated the full-aperture and forward- and backward-looking SLC images from raw data (master and slave).
Produced full-aperture and forward- and backward-looking interferograms via
estimated of the full-aperture interferometric pair’s offset parameters;
co-registries of the slave SLC images; and
a first multi-look processed.
created the full-aperture differential (DInSAR) interferogram and forward- and backward-looking differential interferograms by subtracting a synthetic interferogram simulated from a given DEM the utilized of a full-aperture SAR imaging geometry;
created the forward- and backward-looking residual interferograms with the aid of segment subtraction of the hard-filtered DInSAR interferogram from the forward- and backward-looking interferograms;
generated an MAI interferogram by using phase differencing between the forward- and backward-looking residual interferograms after the second multi-looking and adaptive filtering is utilized for the residual interferograms
generated the final MAI interferogram through a residual phase correction and final DInSAR by the second multi-looking and the adaptive filtering
Figure (2.8): The DInSAR and MAI method processing flow chart.
