Methods for Spatial Data Quality of 3D City Models

Abstract

Abstract 3D city models are nowadays used in very different applications. Due to this, the commercial exchange and, thus,the demand for high quality data becomes more and more important. We describe a quality model that definescommonparametersusedforspatialqualitymeasurement,especiallywhendealingwith3Dcitymodels.Therefore,we explore different representations of city models: the reality, the user’s idea and the digital data set. A well-defined mathematical formalism which addresses the different quality parameters is presented. This formalismalso helps to create algorithms for measurement and improvement of spatial data quality. The aim is to be ableto define uniformed criteria which can be easily transformed into software. The implemented prototype serves asbasefortheevaluationinwhichweprovideexamplesbasedonactualdatasetsusingtypicalqualityrequirements. Categories and Subject Descriptors (according to ACM CCS) : I.3.6 [Computer Graphics]: Methodology andTechniques 1. Introduction 3D city models are used in a large number of practical appli-cations, where high quality data has become indispensable.For example, when city models are – in conjunction withurban development – used to plan new buildings, it is impor-tant to have a data set that is as complete as possible (at leastin the near area around the new buildings). So, the plannercan figure out if the new buildings fit into their surroundingarea or if they cause problems like unexpected shadowing.Another application is the preventive environmental protec-tion, where city models are used to simulate everyday prob-lems like noise pollution and also more heavy ones like floodcatastrophees. In order to create procedures/action plans forflood and noise protection, one needs data sets where objectgeometries are as exact as possible (or at least as exact as itis needed for the specific application).But completeness and exactness of object geometries arenot the only quality characteristics. Think of the applicationof city models in tourism. Tourists, who use a geographicalinformation system to inform theirselves about cities theymay want to visit in the future, do not care about how exactthe 3D coordinates are. Instead, they expect the visualizationof the scene to be close to reality. So, object attributes have tobe accurate, i.e. the building’s color and texture if available.The name and address also have to be correct, as tourists maywant to use a navigation system to find sights in reality later.Moreover, geographical information systems often offer 4Ddata. Therefore, it is important to have a data set with correcttemporal information which is also very up to date.Hence, incorrect data can lead to economic difficulties.Institutions and companies buying spatial data often realizethat there are serious errors in their data sets or that theircity models are out of date. Thus, they spend a lot of moneyand effort to correct these errors, although it would be easierto buy better data sets from 3