Frederik Tack

Airborne photogrammetry and lidar for DSM extraction and 3D change detection over an urban area – a comparative study

A digital surface model (DSM) extracted from stereoscopic aerial images, acquired in March 2000, is compared with a DSM derived from airborne light detection and ranging (lidar) data collected in July 2009. Three densely built-up study areas in the city centre of Ghent, Belgium, are selected, each covering approximately 0.4 km2. The surface models, generated from the two different 3D acquisition methods, are compared qualitatively and quantitatively as to what extent they are suitable in modelling an urban environment, in particular for the 3D reconstruction of buildings. Then the data sets, which are acquired at two different epochs t 1 and t 2, are investigated as to what extent 3D (building) changes can be detected and modelled over the time interval. A difference model, generated by pixel-wise subtracting of both DSMs, indicates changes in elevation. Filters are proposed to differentiate ‘real’ building changes from false alarms provoked by model noise, outliers, vegetation, etc. A final 3D building change model maps all destructed and newly constructed buildings within the time interval t 2 – t 1. Based on the change model, the surface and volume of the building changes can be quantified.

Statistical comparison of urban 3D models from photo modeling and airborne laser scanning

Since a few years, structure from motion has been commercially applied and interesting 3D models are discussed in recent literature. Structure from motion is known to be a relatively fast and low-cost technique for the generation of photo-realistic 3D models. This technique uses a series of digital images taken from different positions. In this image series, each part of the object is recorded on at least three images. Combining these images with information from a metadata file allows to determine the different camera positions and to calculate a textured mesh or colored point set in 3D, by solving a system of geometric matrices. In this article, the geometric quality of urban airborne digital surface models is assessed. These digital surface models are generated by structure from motion. A statistical comparison is performed for an urban area between models generated by structure from motion and models generated by a high density reference point set, acquired by airborne laser scanning. The ease of processing the images in professional software allows to consider structure from motion as an alternative for conventional digital photogrammetry. Although the results for the visual representations of the objects are good, a research on the geometrical accuracy of the final products is discussed in this paper. The statistical analysis of the structure from motion results and the digital surface models generated by other techniques, enables a thorough comparison of the different techniques and the influence of terrain dependent and terrain independent control points.

Measuring and modeling urban dynamics: impact on quality of life and hydrology

The objectives and methodology of the MAMUD research project are presented in this paper. MAMUD is an acronym for Measuring and Modeling Urban Dynamics: Impact on Quality of Live and Hydrology. The research will be conducted over a four year period (2007-2011). The major goal is to investigate how earth observation can contribute to a better monitoring, modeling and understanding of urban dynamics, and its impacts on the urban and suburban environment.

A mixed spaceborne sensor approach for surface modelling of an urban scene

Three-dimensional (3D) surface models are vital for sustainable urban management studies, and there is a nearly unlimited range of possible applications. Along- or across-track pairs from the same set of sensor imagery may not always be available or economical for a certain study area. Therefore, a photogrammetric approach is proposed in which a digital surface model (DSM) is extracted from a stereo pair of satellite images, acquired by different sensors. The results demonstrate that a mixed-sensor approach may offer a sound alternative to the more established along-track pairs. However, one should consider several criteria when selecting a suitable stereo pair. Two cloud-free acquisitions are selected from the IKONOS and QuickBird image archives, characterized by sufficient overlap and optimal stereo constellation in terms of complementarity of the azimuth and elevation angles. A densely built-up area in Istanbul, Turkey, covering 151 km2 and with elevations ranging between sea level and approximately 160 m is presented as the test site. In addition to the general complexity of modelling the surface and elevation of an urban environment, multi-sensor image fusion has other particular difficulties. As the images are acquired from a different orbital pass, at a different date or instant and by a different sensor system, radiometric and geometric dissimilarities can occur, which may hamper the image-matching process. Strategies are presented for radiometric and geometric normalization of the multi-temporal and multi-sensor imagery and to deal with the differences in sensor characteristics. The accuracy of the generated surface model is assessed in comparison with 3D reference points, 3D rooftop vector data and surface models extracted from an along-track IKONOS stereo pair and an IKONOS triplet. When compared with a set of 35 reference GPS check points, the produced mixed-sensor model yields accuracies of 1.22, 1.53 and 2.96 m for the X, Y and Z coordinates, respectively, expressed in terms of root mean square errors (RMSEs). The results show that it is feasible to extract the DSM of a highly urbanized area from a mixed-sensor pair, with accuracies comparable with those observed from the DSM extracted from an along-track pair. Hence, the flexibility of reconstructing valuable elevation models is greatly increased by considering the mixed-sensor approach.

QUANTITATIVE MODELLING OF URBAN CHANGES USING MULTI-TEMPORAL DIGITAL ELEVATION MODELS

The construction of multi-temporal data sets for the modelling and documentation of urban environments has gained a large interest in the last few years. The growing availability of remote sensing data and sophisticated software tools has enabled the construction of Digital Elevation Models (DEMs) with various spatial and temporal resolutions. For this research, multiple scanned airborne images of the inner city of Ghent (Belgium) were processed for the calculation of DEMs using a conventional digital photogrammetric workflow. The aerial images were acquired during four campaigns: 1965, 1977, 1987 and 1990. All resulting image-based DEMs were compared with a DEM acquired with Airborne Laser Scanning (ALS) from 2009. This comparison allowed a model adjustment by minimizing the systematic shift between the data sets. In order to distinct built-up, destroyed or unchanged buildings over time, a threshold of 2.5 m was applied on the resulting vertically shifted points. Finally, a connected component analysis allowed the removal of outliers in the data. The resulting points were evaluated against a 2D digital cadastre map, which enabled a quantitative determination of difference in urban topography. The procedure to detect these changes, as well as the potentials and challenges of this technique, are discussed in this contribution.