Using image contour recognition in GIS navigation

Abstract

This paper presents an approach of using image contour recognition in the navigation for enterprise geographic information systems (GIS). The extraction of object mark images is based on some morphological structural patterns which are described by morphological structural points, contour property, and other geometrical data in a binary image of enterprise geographic information map. Some preprocessing methods, contour smooth following, linearization and extraction patterns of structural points, are introduced. If a focus point of a map is selected, based on the extracted object mark image, a dynamic layout and navigation diagram is constructed, which consists of three underlying views: a graph navigation view, an image layout view, and an enterprise information view. In the diagram each object mark image is represented by a node, the relation information between two nodes is represented by an edge. If a node is selected, the new dynamic layout and navigation diagram is found based on the extracted object mark images around the selected node, which contains the relation with the previous diagram. Therefore a dynamic layout adjustment and navigation for enterprise GIS is determined.