From the street to the metropolitan region: Pedestrian perspective in urban fabric analysis

Abstract

The urban fabric is a fundamental small-scale component of urban form. Its quantitative analysis has so far been limited either in its geographical extent or in the diversity of components analysed. Moreover, the planning approach has traditionally privileged an aerial perspective. A new approach integrating the pedestrian point of view is proposed. Spatial analysis procedures are implemented with a twofold objective: identifying urban fabrics and studying their spatial organization within a large metropolitan area. The former is achieved through multiple fabric assessment, a three-step protocol using a network-based partition of urban space: (i) a set of skeletal streetscape indicators is implemented on each spatial unit, considering different constituents of the urban fabric; (ii) spatial patterns on the street network are identified, applying geostatistical analysis to each indicator; and (iii) spatial patterns with Bayesian clustering are recombined, allowing the identification and characterization of urban fabric types and subspaces within the city. This methodology is tested on the French Riviera metropolitan area, where nine families of urban fabric are identified. Disentangling the spatial organization of urban fabrics represents the second objective of this paper: the geographical distribution of urban fabrics is investigated, applying mathematical morphology and variography while considering networkconstrained topological contiguities.

momepy: Urban Morphology Measuring Toolkit

“…Urban morphology is the area of urban studies that explores the physical form of cities in space and the way it changes in time in relation to the agents of such change (Kropf, 2017). The discipline is based on the analysis of space, traditionally mostly visual and qualitative (Dibble et al, 2015); its objects are the fundamental elements of urban form (building, plot, street) (Moudon, 1997) as well as a range of analytical constructs such as axial maps (Ariza-Villaverde, Jiménez-Hornero, & Ravé, 2013) or proximity bands (Araldi & Fusco, 2019). The increased availability of morphological data and computational power have led in time to more emphasis on quantitative forms of analysis, and the emergence of Urban Morphometrics (UMM) (Dibble et al, 2017): this approach describes urban form via the systematic and comprehensive measurement of its morphological characters.…”

The Character of Urban Japan: Overview of Osaka-Kobe’s Cityscapes

“…Families are obtained through an automated segmentation of the dataset using a naive Bayesian classifier [38]. The best solution found is made of The second methodological protocol, the urban fabric clustering, is based upon multiple fabric assessment (MFA) developed by Fusco (2017, 2019) [40,41]. This method, specifically conceived for describing urban fabrics from a street-based perspective, has proved successful in the analysis of European metropolitan areas [42,43].…”

“…; (2) built-up morphology, containing indicators such as building footprint coverage ratio and presence of the different building types (obtained through the previous clustering application); (3) network-building relationship with for example the building frequency along the street and the street-corridor effect; (4) site morphology, which contains indicators such as the slope and acclivity. The shares of the previously identified building types replace several indicators of the original method developed by Fusco (2017, 2019) [40,41]. The hybridization of these two methods has been presented in Araldi et al, (2018) [44].…”

A Machine Learning Approach to Study the Relationship between Features of the Urban Environment and Street Value

“…Modern European cities are characterised by the presence of different types of urban fabrics; for example, there exist a dense and irregular one, typical of historic centres, a more spaced out and regular one, common to urban extensions of the XIX and early XX century, and a more discontinuous one, characterised by free-standing buildings surrounded by lawns and parking lots, typical of the modernist period. Quantitative measures of the forms of urban fabrics can be found, for example, in the works by Berghauser-Pont and Haupt [31], Gil et al [32], Araldi and Fusco [33], and Venerandi et al [34]. For example, Araldi and Fusco proposed the Multiple Fabric Assessment method (MFA), a technique that automatically detects types of urban fabrics based on morphometric descriptors of proximity bands around street segments, by taking into account the contextual role of interconnected street segments.…”

“…This specific dataset not only had georeferenced representations of street segments, but also information on further aspects, such as street width, category (e.g., highway, regional road, local street), and travel directions. BD TOPO data are also necessary to identify and characterise urban fabrics through the MFA protocol previously mentioned [33]. The street dataset of the French Riviera utilised in this analysis dates back to March 2016 and consists of 98,297 street segments, with a median length of 78 m and a total length of 12,872 km.…”