Comparing intellectual and graphical complexities of traditional and image-derived landuse maps

Remotely sensed images have become a powerful source for landuse mapping. So far, no quantitative comparison and evaluation exist on the differences between an image-derived landuse map and a traditional landuse map. The comparison and evaluation may indicate the possibility for the replacement of a traditional landuse map by an image-derived map. Map complexity is widely used to describe cartographic representations and map effectiveness from intellectual aspect and graphical aspect. This paper quantifies intellectual and graphical map complexities to explore the differences between these two kinds of maps. Intellectual complexity concerns the meaning or significations contained on a map. Graphical complexity concerns spatial characteristics of the graphical content on a map. Results show that the high graphical complexity of the image-derived landuse map is not harmonized with its low intellectual complexity. The intention of this paper is to encourage realistic cognition of the accuracy and problems existing in image-derived landuse maps.