A Line-Based 3d Roof Model Reconstruction Algorithm: Tin-Merging and Reshaping (Tmr)

Abstract

ABSTRACT:Three-dimensional building model is one of the major components of a cyber-city and is vital for the realization of 3D GIS applications. In the last decade, the airborne laser scanning (ALS) data is widely used for 3D building model reconstruction and object extraction. Instead, based on 3D roof structural lines, this paper presents a novel algorithm for automatic roof models reconstruction. A line-based roof model reconstruction algorithm, called TIN-Merging and Reshaping (TMR), is proposed. The roof structural line, such as edges, eaves and ridges, can be measured manually from aerial stereo-pair, derived by feature line matching or inferred from ALS data. The originality of the TMR algorithm for 3D roof modelling is to perform geometric analysis and topology reconstruction among those unstructured lines and then reshapes the roof-type using elevation information from the 3D structural lines. For topology reconstruction, a line constrained Delaunay Triangulation algorithm is adopted where the input structural lines act as constraint and their vertex act as input points. Thus, the constructed TINs will not across the structural lines. Later at the stage of Merging, the shared edge between two TINs will be check if the original structural line exists. If not, those two TINs will be merged into a polygon. Iterative checking and merging of any two neighboured TINs/Polygons will result in roof polygons on the horizontal plane. Finally, at the Reshaping stage any two structural lines with fixed height will be used to adjust a planar function for the whole roof polygon. In case ALS data exist, the Reshaping stage can be simplified by adjusting the point cloud within the roof polygon. The proposed scheme reduces the complexity of 3D roof modelling and makes the modelling process easier. Five test datasets provided by ISPRS WG III/4 located at downtown Toronto, Canada and Vaihingen, Germany are used for experiment. The test sites cover high rise buildings and residential area with diverse roof type. For performance evaluation, the adopted roof structural lines are manually measured from the provided stereo-pair. Experimental results indicate a nearly 100% success rate for topology reconstruction was achieved provided that the 3D structural lines can be enclosed as polygons. On the other hand, the success rate at the Reshaping stage is dependent on the complexity of the rooftop structure. Thus, a visual inspection and semi-automatic adjustment of roof-type is suggested and implemented to complete the roof modelling. The results demonstrate that the proposed scheme is robust and reliable with a high degree of completeness, correctness, and quality, even when a group of connected buildings with multiple layers and mixed roof types is processed.

Automatic Building Extraction From LIDAR Data Covering Complex Urban Scenes

“…In Area 4, in most cases the proposed method offered 2 to 10% better correctness and completeness than Rau (2012), which showed 2% more pixel-based correctness than the proposed method (80 vs 78%). The proposed method also showed better planimetric and height accuracy in this area.…”

“…In all test areas, while the proposed method suffered from more over-segmentation errors, Rau (2012) and Sohn et al (2012) had more under-segmentation errors.…”

“…Extensive and promising experimental results were shown on four data sets. Rau (2012) proposed a line-based roof model reconstruction algorithm, called TIN-Merging and Reshaping. Awrangjeb and Fraser (2014a) proposed a region-growing method based on a plane-fitting technique using LIDAR point cloud data.…”

3D Building Roof Modeling by Optimizing Primitive’s Parameters Using Constraints from LiDAR Data and Aerial Imagery

“…Rau [45] Images 0.8 0.6 Bulatov et al [8] Images 1.1 0.4 Oude Elbrink and Vosselman [46] LiDAR 0.8 0.1 Xiong et al [47] LiDAR 0.7 0.1 Perera et al [48] LiDAR 0.7 0.1 Dorninger and Pfeifer [16] LiDAR 0.8 0.1 Sohn et al [49] LiDAR 0.6 0.2 Xiao et al [50] LiDAR 0.8 0.1 Awrangjeb et al [14] LiDAR 0.8 0.1 Zarea et al [43] LiDAR 0.9 0.4 Our approach LiDAR + Image 0.6 0.1…”

Integrating Smartphone Images and Airborne Lidar Data for Complete Urban Building Modelling

“…A target based graph matching approach was used to handle both complete and incomplete laser data (Elberink and Vosselman, 2009). Rau (2012) proposed a TIN-Merging and Reshaping roof model reconstruction algorithm. Sohn et al (2012) generalized noisy polylines comprising a rooftop model by maximizing a shape regularity.…”