No abstract

A 3D building model retrieval method using airborne LiDAR point clouds as input queries is introduced. Based on the concept of data reuse, available building models in the Internet that have geometric shapes similar to a user-specified point cloud query are retrieved and reused for the purpose of data extraction and building modeling. To retrieve models efficiently, point cloud queries and building models are consistently and compactly encoded by the proposed method. The encoding focuses on the geometries of building roofs, which are the most informative part of a building in airborne LiDAR acquisitions. Spatial histograms of geometric features that describe shapes of building roofs are utilized as shape descriptor, which introduces the properties of shape distinguishability, encoding compactness, rotation invariance, and noise insensitivity. These properties facilitate the feasibility of the proposed approaches for efficient and accurate model retrieval. Analyses on LiDAR data and building model databases and the implementation of web-based retrieval system, which is available at http://pcretrieval.dgl.xyz, demonstrate the feasibility of the proposed method to retrieve polygon models using point clouds.

<p>Forests have high economic and ecological importance. Forest fires and insects (bark beetles in particular) are important disturbance agents putting at risk forest health (resilience). Accurate tree structure metrics and species information are important parameters for forest resources and inventory management. Yet, in many cases this information is not available with adequate spatial and temporal resolution.</p><p>The 4Map4Health project aims to explore the future multitemporal and multispectral laser scanning data in terms of forest application, especially for mapping of the forest health status, tree species, and forest fire risk. Recent studies indicate that multispectral airborne lidar is a useful and meaningful tool to assess moisture of canopies, which is correlated to forest health and susceptibility to disturbance. By means of multitemporal remote sensing data and machine learning, tree species information at individual tree level will be retrieved. During 2021 and Silvilaser 2021 benchmark event, laser scanning data from various platforms, as well as in situ data, have been collected at one of the test sites in eastern Austria. The preliminary outcomes show the high potential for deriving various forest structure parameters valuable for bark beetle risk assessment in addition to topographic and meteorological parameters. Furthermore, first tests show the high potential of ALS data as reference to train various regression models for the assessment of forest structural parameters from Sentinel-1 time series data with high temporal resolution, which can serve as essential input data within a bark beetle risk assessment framework.</p>