Daniela Poli

Fusion of Digital Elevation Models from Various Data Sources

In recent years, collection and processing techniques for Digital Elevation Models (DEMs) generation have improved rapidly, allowing surfaces to be represented with more detail and accuracy. Fusion of overlapping DEMs, generated from data captured with different acquisition techniques, or at different times, allows to find inconsistencies, improve density, accuracy and currency, and eliminate gaps. This is of crucial importance for the improvement of global DEMs, like SRTM (Shuttle Radar Topography Mission). Since the DEMs may have substantial differences, simple amalgamation of all available points would not be satisfying and it would degrade the accuracy of the merged model. Any integration approach aiming at high-quality models needs an increased level of robustness. Computational efficiency and global convergence are further preferable properties. In this paper, an approach is presented for DEM fusion. The goal is to use existing DEMs to create automatically a new DEM which is: geometrically accurate by depicting the correct height information of the area, clean by eliminating blunders and errors which are present in the initial data and complete by modelling all the area on the highest possible resolution. The method is presented and the first results achieved after fusing a Lidar-based DEM and an IKONOS-based DEM on Thun, Switzerland, are shown and commented.

3D modeling of large urban areas with stereo VHR satellite imagery: lessons learned

This paper discusses the potentials of very high-resolution (VHR) stereo imagery for automatic generation of digital surface models (DSM) and 3D information extraction on large metropolitan cities. Stereo images acquired by GeoEye-1 on Dakar (Senegal) and Guatemala City (Guatemala) and by WorldView-2 on Panama City (Panama), Constitucion (Chile), Kabul (Afghanistan), Teheran (Iran), Kathmandu (Nepal), and San Salvador (El Salvador) were processed following a rigorous photogrammetric approach. The work focuses on evaluating the quality of the DSMs in relation to the image and terrain characteristics. The size of the datasets, the variety of case studies, and the complexity of the scenarios allow to critically analyze the potentials of VHR stereo imagery for 3D landscape modeling for natural hazards assessment.

Georeferencing of multi-line CCD array optical sensors with a general photogrammetric model

This paper describes a general sensor model for the georeferencing of imagery from CCD linear array sensors with along-track stereo viewing. The software combines the standard photogrammetric triangulation with a sensor external orientation modeling and self-calibration. The sensor attitude and position, which are different for each image-line, are modeled with time-dependent piecewise polynomial functions while the systematic errors due to lens distortions are described by additional parameters. In case of sensors whose optical system consists of more lenses, the collinearity equations are extended in order to include additional parameters describing the relative orientations of each lens. Using ground control points and, additionally, tie points (TPs), the functions parameters and the ground coordinates of the TPs are estimated in a least-squares adjustment. The described procedures have been applied to imagery from different CCD linear sensors carried on airplane and satellite. In particular, the results obtained with MOMS-P2 imagery are presented and discussed.

DSM generation from optical and SAR high resolution satellite imagery: Methodology, problems and potentialities

The actual high resolution optical and Synthetic Aperture Radar (SAR) satellite sensors offer interesting potentialities for Digital Surface Models (DSMs) generation. Both optical and SAR imagery are characterized by proper deformations and noise due to the different acquisition geometries and processes, which have to be duly taken into account during the DSM generation procedure in order to fully exploit the aforementioned potentialities. The aim of this work is to evaluate the performances of high resolution optical and SAR imagery for DSMs generation over the same testfield area where a dense network of GCPs and LiDAR DSM are available as ground truth data. The image processing and DSMs generation are carried out with the packages SISAR (Software Immagini Satellitari ad Alta Risoluzione) and SAT-PP (SATellite image Precision Processing) while an additional comparison is performed using PCI Geomatica 2012.

DSM generation over large urban areas from VHR satellite sensors

This paper addresses the modeling of large urban areas using Very High Resoluton (VHR) satellite optical sensors. Stereo images acquired by GeoEye-1 on Dakar and Guatemala City and by WorldView-2on Panama City, Constitucion (Chile), Kabul, Teheran, Kathmandu and San Salvador were processed following a rigorous photogrammetric approach and the Digital Surface Models generated. The quality of the DSMs was evaluated in relation to the image and terrain characteristics. The size of the datasets, the variety of case studies and the complexity of the scenarios allow to critically analyze the potentials of VHR stereo imagery for 3D urban modelling.