Future 3D Urbanism: From UAV Data Modelling to Information Visualization

Future 3D Urbanism: From UAV Data Modelling to Information Visualization

Rossana Estanqueiro, José António Tenedório, Carla Rebelo, Joao Pedro Marques
DOI: 10.4018/978-1-7998-2249-3.ch018
(Individual Chapters)
No Current Special Offers


Urbanism has mainly used 2D data for both the urban analysis and diagnosis and the presentation of proposals for changes in the whole city or parts of the city. Even regarding the production of urban indicators, using, for example, the quantification of the existing green area in relation to the resident population, this practice is regularly based on the area and rarely on volume. This situation is mainly justified by the sluggishness and costs associated with obtaining 3D data. The recent development of data collection technology by unmanned aerial vehicles has triggered a change in this scenario. This chapter presents the UAV data acquisition and processing chain, analyses the positional accuracy of UAV data processing performed with GCP measurements obtained from GNSS, demonstrates how positional accuracy assessment and UAV workflow's quality control are relevant for ensuring the accuracy of derived UAV geospatial products, and demonstrates the usability of 3D models in a theoretical 3D urbanism context.
Chapter Preview


Cities and urban areas can be understood from a physical and material perspective as spaces formed by multiple strata or overlapping layers with various characteristics and attributes that, albeit different, confer on them distinctiveness and uniqueness as if they were living organisms. These overlapping layers or strata can be observed and analysed in various fields (social, demographic, cultural, political, economic, etc.) from the dynamics of interaction and very diverse flows. Nevertheless, the intrinsic physical element corresponds to the urban, built and material spatial expression, characterized by its typologies, shape geometries and land-use and occupation. The physical properties directly associated with these spaces are usually studied and analysed on two-dimensional (2D) surfaces.

“Horizontal” (2D) urban analysis has been explored in “classical” studies of the dialectic of compact versus dispersed (“urban sprawl”) urban models of urban growth. Other approaches to the urban form, developed from a quantitative perspective, have made it possible to translate the fundamental characteristics or elements of urban form into quantifiable attributes by calculating indices or indicators – spatial metrics – of a more or less complex nature. These measures support the analytical study of the 2D urban form to support decision making in regional and urban planning, transport planning, environmental and energy planning, analysis and monitoring of urban-rural sustainability, public health and urban safety, socio-spatial exclusion/inclusion, among others. An example of this quantitative perspective is the metrics applied to the study of (sustainable) urban form based on geometric shape parameters (urban shape analysis) obtained in 2D (Estanqueiro, 2011). In this approach, the characteristics of the urban form are understood to be reduced to their fundamental elements or properties, which identify and come close to the phenomenon or reality to be represented or handled using metrics (quantifiable measures in the form of indices or indicators) with a vital spatial element. These can be addressed separately or combined to support urban analysis and monitor the evolution and change of the urban form considering urban sustainability.

Concurrently, more recent studies have explored vertical and volumetric urban analysis, supported by the potential of producing 3D (three-dimensional) urban models. 3D urban modelling is a high potential source of data for the extraction of parameters of interest for studies on urban sustainability in its different dimensions, from a more detailed analysis (intra-urban scale, block, neighbourhood, urban axial segment or set of buildings or urban elements). In fact, the way of thinking about cities and urban planning has resorted to geospatial data that go beyond the two-dimensional (2D). In this sense, the inherent two-dimensionality of classical urban studies is considered insufficient to understand and analyse the tangible domain of urban spaces. Huang, Liu & Lieske (2019) hold that “urban development is a multi-dimensional process that involves changes in urban form in both the horizontal and vertical dimensions” (2019:1). Thus, the use of three-dimensionality in analyses supported not only by horizontality but also by the vertical dimension and the volumetric arrangement of the set of urban elements and objects is additionally recommended. This idea is backed up by authors that admit the use of 3D data to support the production of relevant and useful geographical information for decision making in urban planning and urbanism (Shiode, 2001; Batty, 2000; Batty et al., 2000; Zhang, Zhu, & Wang, 2004; Yin & Shiode, 2014; Graham & Hewitt, 2012). Graham & Hewitt (2012) crucially argue that “(…) a fully volumetric urbanism is required which addresses the ways in which horizontal and vertical extensions, imaginaries, materialities and lived practices intersect and mutually construct each other within and between subterranean, surficial and suprasurface domains” (2012:74).

Key Terms in this Chapter

UAV Data Processing: Corresponds to the sequential linking of a set of processes based on computer vision and traditional photogrammetry techniques. It consists of the processing of high-overlapping UAV image pairs to generate a 3D model represented by a dense point cloud.

Urban Planning: Is a process supported by politics, technology, techniques and methods, which originates or conditions the physical, environmental and social dynamics and the design of a city.

Positional Accuracy: Measures the spatial quality of geographical data, horizontally and vertically, by comparison with a more accurate geospatial technique, where the statistical root mean square error (horizontal and vertical) is calculated based on the mean of the squared positional errors at checkpoints.

3D Model: Is a digital representation of an object or scene in a three-dimensional reference system. This digital representation is made in two different formats: vector as geometric topological reconstruction using points, lines and areas; or raster in grid cells (voxel). The resolution of the 3D model will depend on the level of detail required in the representation of 3D information.

Urban Data: Are graphic and alphanumeric records with urban spatial reference, considering highly accurate scales of analysis and originating information necessary for understanding, knowledge and decision making in the fields of urbanism and urban planning.

3D UAV Data: Are represented by a georeferenced dense point cloud, obtained from semi-automatic photogrammetric processing of UAV image pairs acquired by a small aerial platform system.

Urban Big Data Processing: Are toolbox and programming models that allow the processing of large-scale data in permanent evolution to extract information and behaviours needed to support decision making in cities.

Complete Chapter List

Search this Book: