Abstract
Abstract. Standard interferometry poses a challenge in non-urban areas due to temporal and spatial decorrelation of the radar signal, where there is high signal noise. Techniques such as Small Baseline Subset Algorithm (SBAS) have been proposed to make use of multiple interferometric combinations to alleviate the problem. However, the interferograms used in SBAS are multilooked with a boxcar (rectangle) filter to reduce phase noise, resulting in a loss of resolution and signal superstition from different objects. In this paper, we proposed a modified adaptive spatial filtering algorithm for accurate estimation of interferogram and coherence without resolution loss even in rural areas, to better support the deformation monitoring with time series interferometric synthetic aperture radar (InSAR) technique. The implemented method identifies the statistically homogenous pixels in a neighbourhood based on the goodness-of-fit test, and then applies an adaptive spatial filtering of interferograms. Three statistical tests for the identification of distributed targets will be presented, applied to real data. PALSAR data of the yellow river delta in China is used for demonstrating the effectiveness of this algorithm in rural areas.
Highlights
interferometric synthetic aperture radar (InSAR) is a microwave remote sensing technique that uses satellite images to measure surface deformation over large areas with millimetre precision (Rosen, Hensley et al, 2000)
With the availability of various synthetic aperture radar (SAR) images and continual observations over the same area, long time deformation can be extracted by using stacking techniques such as Persistent Scatterer Interferometry (PSI) (Ferretti, Prati et al, 2000; 2001) and Small Baseline Subset Algorithm SBAS (Berardino, Fornaro et al, 2002; Mora, Lanari et al, 2002)
While in non-urban areas, characterized by vegetated or low reflectivity homogenous regions, such as cropland, volcanoes, mines, reservoirs, the spatial density of persistent scatterers (PSs) extracted by PSI is low (< 10 PS/sqkm), which is a key limitation for its application in rural areas (Ferretti, Fumagalli et al, 2011)
Summary
InSAR is a microwave remote sensing technique that uses satellite images to measure surface deformation over large areas with millimetre precision (Rosen, Hensley et al, 2000). PSI processes differential interferograms with respect to a common master image, aims to identify coherent targets exhibiting high phase stability during the whole time period of observation These phase stable points, slightly affected by temporal and geometrical decorrelation, are called persistent scatterers (PSs). DS belongs to areas of moderate coherence in some interferometric pairs of stack, where there is sufficiently high number of random small scatterers within a resolution cell with no dominant scatterer, and follows the complex circular Gaussian distribution (Bamler and Hartl, 1998) Various techniques, such as SBAS and SqueeSAR (Ferretti, Fumagalli et al, 2011) have been proposed to process DS, which are widespread in rural areas
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
