Context-based automatic reconstruction and texturing of 3D urban terrain for quick-response tasks.
Abstract: Highly detailed 3D urban terrain models are the base for quick response tasks with indispensable human participation, e.g., disaster management. Thus, it is important to automate and accelerate the process of urban terrain modeling from sensor data such that the resulting 3D model is semantic, compact, recognizable, and easily usable for training and simulation purposes. To provide essential geometric attributes, buildings and trees must be identified among elevated objects in digital surface models. After building ground-plan estimation and roof details analysis, images from oblique airborne imagery are used to cover building faces with up-to-date texture thus achieving a better recognizability of the model. The three steps of the texturing procedure are sensor pose estimation, assessment of polygons projected into the images, and texture synthesis. Free geographic data, providing additional information about streets, forest areas, and other topographic object types, suppress false alarms and enrich the reconstruction results. [Copyright &y& Elsevier]/nCopyright of ISPRS Journal of Photogrammetry & Remote Sensing is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
In another measurement campaign five years later, several UAV-videos covering building walls were recorded. A video sequence with around 200 frames was processed by a SLAM method
Presentation of algorithms, testing.
Presentation of a robust, modular algorithm for context-based urban terrain modeling from sensor datawe proposed and described an automatic and a semi-automatic method for pose estimationAlgortihm for: geometric reconstruction, roof analysis and visibility analysis and texturing processes, to synthesise data
The input for our building detection procedure is an elevation map, or, if geo-referenced, a Digital Surface Model (DSM) that either results from an Airborne Laser Scan (ALS) or is sampled from a dense point cloud resulting from one or several depth maps.The first data set we present stems from the village Bonnland, a widely used urban training facility in Southern Germany.The input data is a DSM computed from an ALS point cloud of as well as a corresponding digital orthophoto. The 3D building polygons were generated by methods of Gross et al. (2005).The second data set (Vaihingen area 3) is an ISPRS WG III/4 benchmark due to Rottensteiner et al. (2012) for urban terrain reconstruction. It represents a purely residential area in the city of Vaihingen (Germany) and contains 56 small, detached houses and many trees.the last data set is the area 4 of the Toronto data set which is also an ISPRS benchmark.
However, this procedure for reconstructing roofs has problems if there are many flat roofs and many outliers in the data.
We presented a robust, modular algorithm for context-based urban terrain modeling from sensor data. Except for the choice of parameter values and integration of additional sources of information, the algorithm is automatic and requires only a 2.5D elevation map as input.The elevation map may stem from an airborne laser scan or may be computed by a dense matching algorithm
The purpose is to demonstrate that an existing model can be updated, enriched by new information, and exported into simulation software.
 |
Il sito web Portfolio of Solutions è stato inizialmente sviluppato nell'ambito del progetto DRIVER+. Oggi, il servizio è gestito da AIT Austrian Institute of Technology GmbH, a beneficio della gestione europea delle . PoS è approvato e supportato dal Disaster Competence Network Austria (DCNA) così come dai progetti STAMINA e TeamAware H2020. |