| dc.description.abstract | The cyber city has demonstrated its potential as a replica of the real one in urban and environmental planning, design, construction, and management. The building model is one of the most important elements in a cyber city. Traditionally, the reconstruction of building models is performed by using aerial photography. An emerging technology, the airborne lidar (Light Detection and Ranging) system provides a promising alternative. Hence, in this investigation we utilize lidar point clouds for building reconstruction.
The first part of this investigation presents a scheme for the reconstruction of building models from lidar point clouds and topographic maps using the divide-and-conquer strategy. The proposed scheme comprises three major parts: (1) decomposition of building boundaries; (2) shaping of building primitives; and (3) combination of building primitives. In the decomposition of building boundaries, the lidar data is selected to extract the inner structure lines. Then, building boundaries are divided using the extracted feature lines by the split procedure into several building primitives. To shape the building primitives, parameter fitting is applied to shape the roof for each building primitive from lidar point clouds. The roof shapes include both planar and circular types. Finally, a least squares adjustment process which considers the co-planarity and co-linearity is used to merge the 3-D building primitives into building models.
In the second part of this investigation the effects of point cloud density for roof splitting and roof shaping are analyzed. Since the lidar is a non-targeting sampling system, the measurements are randomly distributed over the surface. Thus, the density of point clouds is an important issue in the reconstruction of complex objects. We focus on the relationship among point density, noise level, roof complexity, and the accuracy of generated roofs. Experimental results indicate that the accuracy improves as the point density increases. In shaping accuracy results, an accuracy of 15cm may be reached when the outliers are smaller than 30%. For non-flat roofs, the same accuracy may be achieved, provided that no more than at most 15% outliers exist.
The proposed method is tested with the data collected from Taipei and Pingdong city in Taiwan. The reconstruction rate is better than 90% while the omission error is smaller than 5%. The planimetric and vertical accuracy of the reconstructed models are both better than 50cm. The experimental results confirm that the proposed scheme produces high fidelity models. | en_US |