Multi-INT fusion of GEOINT and IMINT can enable performance optimization of target detection and target tracking problem domains1, amongst others. Contextual information, which defines the relationship of foreground to background scene content, is a source of GEOINT for which various online repositories exist today including but not limited to the following: Open Street Maps (OSM)2 and the United States Geological Survey (USGS)3. However, as the nature of the world’s landscape is dynamic and ever-changing, such contextual information can easily become stagnant and irrelevant if not maintained. In this paper we discuss our approach to providing the latest relevant context by performing automated scene generated background context segmentation and classification for near-nadir look angles for the purpose of defining roadways or parking lots, buildings, and natural areas. This information can be used in a variety of ways including augmenting context data from repositories, performing mission pre-planning, and for real-time missions such that GEOINT and IMINT fusion can occur and enable significant performance advantages in target detection and tracking applications in all areas of the world.