Fabio Dell’Acqua

Urban remote sensing using multiple data sets: Past, present, and future

Abstract This presentation aims at stimulating ongoing research and development on the topic of urban remote sensing exploiting multiple data sets by offering a bibliographic review of the state-of-the-art and a look at the future in terms of the identifiable needs and advancement potential. The increasing availability of large amounts of data suitable for urban applications demands the development of adequate algorithms and procedures for their (semi-) automatic characterization. This study takes stock of the current achievements as reported in the technical literature that are particularly relevant to the interests of the remote sensing & information technology (RS&IT) community. It is also aimed at foreseeing and proposing future lines of research, focusing on the algorithms and techniques particularly suited for multiple data sets analysis in the urban areas. Many sensor types (optical, thermal, laser, SAR) are considered, and particular attention is given to satellite data analysis. After a review of the existing literature, the paper describes interesting application fields that are very much open for further research and the potential that urban remote sensing and urban data fusion offer to end users.

First results on road network extraction and fusion on optical and SAR images using a multi-scale adaptive approach

In this article, we introduce a unitary approach to road extraction in wide area images, obtained by means of satellite sensors in both the optical/infrared and microwave domains. Despite the large amount of methodologies discussed in technical literature for road extraction, they have been mostly tested on relatively small portions of satellite images. Moreover, in many cases, the method targeted an optical or a synthetic aperture radar (SAR) image, and a unitary strategy is missing. This study is aimed at bridging these gaps and provides a unique framework for the extraction of roads with different characteristics using optical or SAR data sets. The approach exploits a multi-scale analysis to adapt to the different resolutions of data and a pre-processing step to adapt to the different wavelengths of data. When possible, the framework allows the fusion of the road networks extracted from optical and SAR data of the same area. The soundness of the approach is proved by means of the analysis of Landsat and ALOS data of an area in Congo.

Integration of EO-based vulnerability estimation into EO-based seismic damage assessment: a case study on L’Aquila, Italy, 2009 earthquake

Remote sensing is proving very useful for identifying damage and planning support activities after an earthquake has stricken. Radar sensors increasingly show their value as a tool for damage detection, due to their shape-sensitiveness, their extreme versatility and operability, all weather conditions. The previous work of our research group, conducted on 1-m resolution spotlight images produced by COSMO-SkyMed, has led to the discovery of a link between some selected texture measures, computed on radar maps over single blocks of an urban area, and the damage found in these neighbourhoods. Texture-to-damage correlation was used to develop a SAR-based damage assessment method, but significant residual within-class variability makes estimations sometimes unreliable. Among the possible remedies, the injection of physical vulnerability data into the model was suggested. The idea here is to do so while keeping all the sources of data in the EO domain, by estimating physical vulnerability from the observation of high-resolution optical data on the area of interest. Although preliminary results seem to suggest that no significant improvement can be directly obtained on classification accuracy, there appears to be some link between estimated damage and estimated accuracy on which to build a more refined version of the damage estimator.

Rapid Mapping Using Airborne and Satellite SAR Images

Historically, Synthetic Aperture Radar (SAR) data was made available later than optical data for the purpose of land cover classification (Landsat Legacy Project Website, http://library01.gsfc.nasa.gov/landsat/; NASA Jet Propulsion Laboratory: Missions, http://jpl.nasa.gov/missions/missiondetails.cfm?mission

Fusion of Optical and SAR Data for Seismic Vulnerability Mapping of Buildings

=

Change Detection in Urban Areas: Spatial and Temporal Scales

Seasat); in more recent times, the milestone of spaceborne meter resolution was reached by multispectral optical data first (Ikonos; GEOEye Imagery Sources, http://www.geoeye.com/CorpSite/products/imagery-sources/Default.aspx#ikonos), followed a few years later by radar data (COSMO/SkyMed [Caltagirone et al. 2001] and TerraSAR-X [Werninghaus et al. 2004]). As a consequence, more experience has been accumulated on the extraction of cartographic features from optical rather than SAR data, although in some cases radar data is highly recommendable because of frequent cloud cover (Attema et al. 1998) or because the information of interest is better visible at the microwave frequencies rather than at the optical ones (Kurosu et al. 1995).

Comparison of efficient sparse reconstruction techniques applied to inverse synthetic aperture radar images

Seismic risk depends not only on seismic hazard, but also on the vulnerability of exposed elements since it is important in providing the necessary information to policy and decision-makers in order to prevent and mitigate the loss in lives and property. Currently, the estimation of seismic vulnerability of buildings relies on accurate, complex models to be fed with large amounts of in situ data. A limited geographical scope is a natural consequence of such approach, while extensive assessment would be desirable when risk scenarios are concerned. Remote sensing might be fruitfully exploited in this case, if not for a gap between information required by current, accurate, data-hungry vulnerability models and information derivable from remotely sensed data. In this context, naturally the greatest amount of information should be collected, and data fusion is more a necessity than an option. Fusion between optical and radar data allows covering the widest range of information pieces; in this chapter we will describe how such information may be extracted and how it can be profitably fed to simplified seismic vulnerability models to assign a seismic vulnerability class to each building. Some examples of real cases will also be presented with a special focus on the test site of Messina, Italy, a notorious seismic-prone area, where an intensive campaign of data collection is in progress within our research group.

Efficient Geospatial Analysis of Remotely Sensed Images by Means of Linear Feature Extraction and Combination

Urban areas are a challenging environment because of their ever changing structure and the different temporal behaviors and spatial patterns. In this chapter a detailed analysis of some of the questions arising from the use of remotely sensed data in urban area for change detection are addressed. Specifically, the role of very high resolution sensors and their relevance with respect to either fast or slow changes in human settlement is analyzed, with specific stress on rapid mapping in specific sites (hotspots), e.g. for post-disaster damage assessment. Similarly, the possibility to exploit long temporal sequences of coarser resolution data is also explored and discussed, since the availability of huge archives is nowadays a reality that may be used to look for interesting interrelationships between urban area pattern changes and environmental changes, at both the local (town), regional and global level. Examples related to a so-called “hypertemporal” sequences of EO data are offered, and show the great potentials of these data sets.

Post-event Only VHR Radar Satellite Data for Automated Damage Assessment

Abstract. Compressed sensing can be a valuable method with which to acquire high-resolution images, reducing the stored amount of information. This objective may be pursued without using any prior knowledge of the images, unlike the standard information compression algorithms do. Information compression can be obtained by a simple matrix multiplication, but the process of reconstructing the original image could be very expensive in terms of computation requirements. We are interested in comparing different reconstruction techniques for compressed air-to-air inverse synthetic aperture radar images, looking for a sensible compromise between performance results and complexity. In more detail, the compared algorithms are iterative thresholding, basis pursuit and convex optimization. Furthermore, particular attention has been devoted to a more appropriate way of splitting large-sized images in order to obtain smaller matrices with uniform sparseness for reducing the computational load.

Spectral Resolution in the Context of Very High Resolution Urban Remote Sensing

An efficient and valid interpretation of very high resolution images, either SAR or optical, must take into account the spatial details of the scene. As a result, there are many options to the very simple pixel-by-pixel classification scheme which applies to coarse resolution remote sensing data sets. One option is the exploitation of spatial features, like for example textures, that connect each pixel with its neighborhood (Dekker 2003). Another option is multi-scale analysis, where scales, and thus contexts with different size, are jointly considered to capture details at various levels (Benediktsson et al. 2003). A third option, explored in this chapter, is to extract significant yet simple geometrical features and use them (possibly in combination with spectral features) to improve the understanding of the scene (Xin et al. 2007). It is a separate approach, other than standard classification, but it can be used also for classification and change detection (which is essentially multitemporal classification).