Search

This paper describes a new visualization approach for the automatic generation of visual summaries dealing with cartographic visualization methods and modeling of real time data coming from sensors. Indeed the concept of chorems seems an interesting candidate to visualize real time geographic database summaries. Chorems have been defined by Roger Brunet as schematized visual representations of territories. However, the time information is not yet handled in existing chorematic map approaches, that is the issue been discussed in this paper in which geodata are coming regularly from sensors distributed along some territory. Our approach is based on spatial analysis by interpolating the values recorded at the same time, so we have a number of distributed observations on areas of study. To get a better visual overview of the entire sensor geodata at a given time, we use spatial statistics formulas on the fly, and so it is possible to extract important spatiotemporal patterns and detect trends over time as geographic rules. Then, those spatiotemporal patterns are visualized as animated chorems. An example is taken from meteorology. [ABSTRACT FROM AUTHOR]

Abstract: The need to support the activities of decision makers through highly interactive visual environments has motivated the growing interest in the area of GeoVisual Analytics. New interactive visualization tools are being envisaged to deal with large datasets in order to synthesize information and perform complex analytical tasks. Along this line, our research efforts have been focusing on new cartographic approaches which could support daily analysts’ work by producing synthesis and presentation of discovered patterns in a concise and understandable way. As a result, we have proposed the adoption of chorems as an innovative method to visually summarize information from spatial databases and we have implemented an XML-based language, named ChorML, able to both specify chorems characterizing a map and store the information useful to their manipulation. The goal of the present paper is to enhance the role that chorems may play in geographic domains, by exploiting them also for querying and accessing data associated with a phenomenon under investigation. To develop this idea, we first extend the semantics associated with the chorem concept and define a set of operators useful for the rapid analysis of spatio-temporal phenomena. Then, starting from an initial prototype, we present a chorem-based visual environment that integrates traditional interactive visualization and analysis techniques. The environment is specifically conceived so that each visual interaction task has a context-sensitive behavior, which allows users to acquire specific information from the underlying spatial database. Finally, we present an enhanced version of ChorML language, able to support the new analysis functionalities on chorems. [Copyright &y& Elsevier]