Making 21st Century Maps Legible - Methods for Measuring and Improving the Legibility and Usability of Real-Time Maps

Abstract: The research presented in this thesis has focused on methods for improving the legibility and usability of maps in real-time map services. The aim has been threefold: to create methods to increase the cartographic quality of maps in map services, to develop methods to measure the legibility, as perceived by the user, and to evaluate the quality of maps. Map services are becoming increasingly common. This is a result of improved technology and an increasing number of ideas and requests for map applications. Some examples of map applications are: positioning services, navigation services and services for searching, viewing and downloading geospatial information. However, it is a challenge to produce good maps without the input of a cartographer. Therefore, in order to increase the cartographic quality in map services, we need to improve the automatic methods. In this thesis, a framework consisting of appropriate methods, which need to be established, are presented. The methods are divided into five categories: semantic, conflation, generalization, label placement and symbolization methods. This thesis presents studies of the last four categories. Conflation is required for map services that display related information from several sources to avoid inconsistencies. In the work presented in this thesis, conflation methods have been developed in order to integrate routes into topographic maps. The datasets have been merged into one in order to improve the quality of the representation of objects and to resolve geometric inconsistencies. Generalization methods are performed in order to reduce or avoid clutter. In order to guide generalization and achieve legible maps for real-time map services, we need analytical descriptions, or constraints, that describe the legibility of maps. In the work presented in this thesis, map legibility measures have been divided into four types: amount of information, spatial distribution, object complexity and graphical resolution. A number of measures of each category, and methods for describing synthesis of measures, have been evaluated. The results showed that legibility measures and syntheses of these can provide valuable help in describing map legibility. Label placement methods concerns text and icons. In the work presented in this thesis a method for performing icon placement in real-time has been developed. With this method, icons are placed in positions where they obscure as little map information as possible. Good label placement like this aids the map reader in interpreting information provided by the map. Symbolization methods are essential to ensure good representation of geographic data from different sources. In the work presented in this thesis different symbolizations have been evaluated in order to find the most appropriate type to use for maps in map services. Map services require symbolization methods that can provide appropriate colours and good contrast between symbols from different datasets in real-time. The correct use of symbology allows visual hierarchies to be created, more detailed geometric information to be discerned and different object types to be discriminated. The work presented in this thesis has been performed with the overall aim to increase the usability of the maps in map services. A number of different user tests have been performed in order to evaluate the legibility and usability of maps. Usability focuses on the user rather than the product, arguing that not only is it important that the product itself functions well, but it must function well in its interaction with the user. In order to achieve this, a user-centred approach is invaluable, also concerning the design and development of maps and map services.