StreetExplorer: Visual Exploration of Feature-based Patterns in Urban Street Networks

The analysis of street networks is an important problem in applications like city planning, comparison of urban street properties, or transportation network analysis. Graph-theoretic computation schemes today provide street network analysts with a range of topological features relating e.g., to connectivity properties of street networks. Typically, an abundance of different network features is available, and some or all of these features may be relevant for within- and between comparison tasks at different scales across the network. Therefore, it is desirable to interactively explore the large segment feature space, with the goal of finding interesting patterns based on extracted features, taking into account also the geospatial properties of a given network. We introduce StreetExplorer, an interactive visualization system for the exploration of global and local properties of urban street networks. The system is based on a set of appropriate similarity functions, which take into account both topological and geometric features of a street network. Together with a set of suitable interaction functions that allow the selection of portions of a given street network, we support the analysis and comparison of street network properties between and across features and areas. We enhance the visual comparison of street network patterns by a suitable color-mapping and boosting scheme to visualize both the similarity between street network portions as well as the distribution of network features on the segment level. Together with experts from the urban morphology analysis domain, we apply our approach to analyze and compare two urban street networks, identifying patterns of historic development and modern planning approaches, demonstrating the usefulness of StreetExplorer.