Maps and charts in the web, whether as images or embedded in digital documents, contain valuable information that is not fully available to information retrieval tools (search engines). The information used to describe these maps (charts) is typically extracted from the (image) metadata rather than from the information the graphic was initially designed to express. Creating metadata for a map (chart) is a manual task that only a small number of users are good at. The idea is to automatically read the map (chart) data and provide the user with a textual summary of the map (chart).