DBPapers
DOI: 10.5593/sgem2017/23/S10.010

DIFFERENCES IN THE QUANTITY AND DISTRIBUTION OF THE PS CANDIDATES IN VV AND VH SENTINEL-1 DATA SET

H.Malik, A. Lesniak, R. Murdzek
Monday 11 September 2017 by Libadmin2017

References: 17th International Multidisciplinary Scientific GeoConference SGEM 2017, www.sgem.org, SGEM2017 Conference Proceedings, ISBN 978-619-7408-03-4 / ISSN 1314-2704, 29 June - 5 July, 2017, Vol. 17, Issue 23, 71-78 pp, DOI: 10.5593/sgem2017/23/S10.010

ABSTRACT

Satellites constellation Sentinel-1 provide costless SAR (Synthetic Aperture Radar) images of global range. Scientific institutions operating with SAR data received the great opportunity for carrying out continuous observations of the Earth. This kind of policy affects the efficiency of performed research and has direct impact on the safety in case on environmental threats. It is possible to monitor ground deformations of endangered areas. Sentinel-1 datasets have great potential in PSInSAR (Persistent Scatterer Interferometric SAR) technique. Short revisit time allows for generating submillimeter subsidence maps with global range. That image acquisition frequency cause a decrease of temporal decorrelation which is crucial in that kind of analysis. It is important to monitor areas with mining activities (like Upper Silesian Coal Basin in Poland). Early detected subsidence can help prevent damage of buildings. In this paper, differences in quantity and distribution of PS (Persistent Scatterer ) candidates in VV and VH polarizations were investigated. Amplitude dispersion was used for PS candidates selection. PS candidates were selected in three variants by thresholding amplitude dispersion at 0.2, 0.3 and 0.4 level. PS candidates density map according to the type of the ground was shown. Heavily built-up, vegetated and intermediate areas were specified. Combining both polarization in the analysis one can increase amount of final PS points, resulting in more comprehensive and more accurate results. Number of PS candidates points that exist in both data sets were examined and subjected to pointwise analysis. This work is preliminary research conducting to new dual-pol- PSInSAR technique using both polarization of Sentinel-1 data sets. For this study thirteen Sentinel-1A images between 11th January and 26th August were used. Final results were presented in the example of Cracow (city in Poland).

Keywords: PSInSAR, Sentinel-1A, VV and VH polarizations