LiDAR digital building models for true orthophoto generation

摘要

The importance of digital orthophotos in spatial\uddatabases has increased in recent years, since they are an\udefficient, low-cost and, if properly managed, accurate\udproduct. Usually, the generation of orthophotos is carried\udout using digital terrain models (DTMs); meaning without\udtaking into account vegetation, buildings, and other\udattached and detached structures. This leads to low\udaccuracies in urban areas, bringing distortions into the\udimage. To avoid this unwanted effect, one must adopt a\uddigital surface model (DSM), as proposed by Amhar et al.\ud(Int Arch Photogrammetry Remote Sens 32(4):16–22,\ud1998). The method proposed in this paper allows for the\udcreation of true orthophotos by using a DSM to refine the\udrepresentation of buildings. The pixel size of the DSM must\udbe similar to that of the true orthophoto in order to model\udthe roof edges with sufficient accuracy. This paper presents\uda new method capable of correcting the roof displacement\udusing an approach based on the integration of several\udproducts today available in public administrations, such as\uda geodatabase, DTMs/DSMs, and light detection and\udranging (LiDAR) data. The method is based on a rigorous\udmodelling of simple roofs starting from their 2D projection\udin the geodatabase, while information about their heights\udcan be obtained using LiDAR data. For some selected\udsimple roofs, automatic modelling can be carried out, in\udwhich a robust interpolation method, such as RANSAC, is\udused to model the pitches identified by a clustering\udprocedure. For complex roofs, where creating a rigorous\udmodel in a fully automatic way is not possible, a procedure\udbased on the thickening of a DSM is carried out.