Exploiting Raster Maps for Imagery Analysis

Maps are an incredibly rich source of information, but the information contained in a map is often locked up in a raster image. Sometimes the original source data used to create the map is available, but more often than not this information has either been lost or is not available. In this project, we plan to develop the technology that will make it possible to exploit the huge number of raster maps available and use them for imagery analysis. In particular, we will develop the technology for automatically finding online raster maps, automatically aligning the raster maps with satellite imagery, and automatically extracting the information contained in a map, such as transportation networks, hydrographic layers, and the textual labels on the information. The resulting technology will allow an analyst to view a satellite image for any place in the world, automatically find and align the maps covering that region, and then overlay selected layers from the map to better understand the information shown in an image.