FIGURE 3.1 The complexity and methods used in GEOINT fusion depend both on the size of the geographical area (horizontal axis) and the length of the time period (vertical axis) being covered. In this figure, the classification of use cases is shown by these dimensions.
may be georegistered to a reference map by aligning specific image pixels to corresponding map landmarks. At level 1 (object/entity), information from multiple sensors with overlapping sensing ranges is combined to estimate properties (e.g., location, shape, size, type) of an identifiable entity, such as a vehicle or building. For example, a national air-traffic monitor room may track every aircraft using information collected from local air-traffic controllers. At level 2 (situation assessment), information from all sources is combined to estimate the impact of a recent event or behavior on a geographic area of interest. For example, an emergency manager may fuse weather prediction data sets, plume simulation maps, population density maps, and transportation maps to identify emergency evacuation routes.
The subobject, object, and situation assessment levels are often concerned with a single point in time (snapshot). However, multiple time frames can be considered at any level. At level 3 (impact assessment), a recent image may be compared with an older image to detect major changes in an object or geographic area of interest. For example, the impact of a forest fire may be assessed by comparing remotely sensed images before and after the fire. At level 4 (process refinement), the process of data collection and fusion is refined using what could be considered “control law” that depends on a utility function expressing the dependence of fusion quality on input quality. For example, fusion may be used to reconfigure the locations and trajectories of sensor platforms (e.g., satellites, aircraft, vehicles) to closely monitor an event (e.g., hurricane) or high-value target in order to improve the quality of fused output estimates of interest.
The late 1990s brought the establishment of the International Society for Information Fusion as well as two journals dedicated to information fusion: Journal of Advances in Information Fusion and Information Fusion. Conference discussions and publications have refined the use cases in new directions. For example, a long time series of snapshots may enrich traditional fusion with concepts from time-space geography ( Hägerstrand, 1967) and the dynamics of geographic domain ( Hornsby and Yuan, 2008), leading to a new use case (level 5). At the location (e.g., pixel) geographic footprint, a past time series of measurements can be used to determine a statistical distribution, which, in turn, can be used to evaluate future values for anomalies, regime-change points, and other factors. At an identifiable-entity geographic footprint, a time series of locations produced by multiple sensor measurements for an object can be fused to estimate the object’s trajectory, which can be processed further to identify its frequent locations, routes, schedules, and other spatiotemporal patterns (Shekhar et al., 2011).