Road Grid Extraction and Verification

While maps exist for most urban areas, there are many locations where the information is not accurate, it may be out of date, or it may be incomplete or of insufficient resolution for the applications. Many difficult problems remain in automated cartography. One of them is the extraction of a street grid in an urban environment. Much of the work on road detection has concentrated on either low resolution, primarily rural roads (usually producing “spaghetti” roads with no notion of intersections) or high resolution roads without the topological information of the intersections. This paper address the problem of extracting a grid with the topological information intact. Given an initial seed intersection, which gives the size and orientation of the regular grid, this system uses a feature-based hypothesis and verify paradigm to find the street grid. Verification uses local context, provided by an intersection model and by an extended street model, and any available sensors.