DInSAR technique has been long used to observe and monitor ground surface changes over large areas. Multi- temporal SAR image datasets are used for such investigations. To eliminate the issues of signal decorrelation associated with DInSAR based processing, Persistent Scatterer InSAR (PSInSAR) technique has come in a big way in the recent years where deformation studies at small scales and fine accuracy (mm-level) have been at- tempted and measured. In this paper, ground deformation of the area near Naples, Italy has been estimated for the year 2014 using a spatial correlation based PSInSAR method. Here, Cosmo-SkyMed (CSK) SLC (ascending orbit, single polarisation, stripmap Himage mode) images of Very High Resolution (VHR) were used in this study due to its capability to detect small-scale ground deformation signal over an urban area. The PSInSAR processing, used here, involves two stage selection of PS points or stable scatterers, with the coregistered SLCs and differential interferograms, using amplitude and phase analysis. The PSInSAR results were validated using time series data of two continuous GPS (cGPS) stations over the same period for the study area. The mean deformation rate over the study area was observed to be varying from -15 to +18 mm/year along line-of-sight (LOS) of Cosmo-SkyMed. Comparison between PSI derived deformation time series and cGPS measurements re- veal a good correlation with minimal discrepancies. Additionally, distinct differences were observed between the PS based LOS displacements with those obtained from cGPS based observations in case of one station compared to the other.
The goal of our work is to use visual attention to enhance autonomous driving performance. We present two methods of predicting visual attention maps. The first method is a supervised learning approach in which we collect eye-gaze data for the task of driving and use this to train a model for predicting the attention map. The second method is a novel unsupervised approach where we train a model to learn to predict attention as it learns to drive a car. Finally, we present a comparative study of our results and show that the supervised approach for predicting attention when incorporated performs better than other approaches.