A primary cause of failure in tactical air-to-ground missions is the the aircrew fails to find the assigned target. This is the case even for missions against preassigned, large, stationary targets, such as bridges, supply depots, and air-defense sites, for which reconnaissance and satellite images of the target and its surroundings are available for the mission planning process. An important reason for this poor performance is the limited time available for target acquisition, especially when target defences necessitate low-level ingress. During the time a target is visible to aircraft sensors, the aircraft is also visible to the target's defense systems. The time required for air-to-ground targeting with present systems (for example, to lock a targeting FLIR onto an aimpoint is comparable to the response time of air defense systems). Clearly, minimizing the time needed to find and recognize targets is crucial for making attack missions more survivable. Survivability is enhanced if the attack mission can be carried out without the need to 'pop up' to use a targeting sensor. In a typical low-level attack mission, 8-10 s is available for the targeting process, from appearance of the target to weapon release. However, a pilot cannot remove his attention from flying the aircraft for this length of time, especially at night using night-vision systems. Pilots who have tested night-vision systems for low-level flight estimate that 1.5-2.0 s is the maximum allowable time to concentrate on a head-down display under these conditions. A near-term solution to the targeting problem must let the human operator make the final decision. The system described in this paper automates the search for the target (or an offset aimpoint) by sending video from one or more of the aircraft sensors to an image correlator. The correlator finds the best match to an image of the target area, supplied from mission planning data, and displays the resulting target position overlaid on the sensor image. The aircrew is required to look at a head-down display only long enough to confirm of deny the correlator's result. The correlator can then automate the process of locking an image missile seeker onto target.
Multispectral Image Correlation for Air-to-Ground Targeting
1995
12 pages
Report
Keine Angabe
Englisch
Multispectral Image Correlation for Air-to-Ground Targeting
British Library Conference Proceedings | 1995
|NTRS | 1977
|Multispectral-scanner image processing
NTRS | 1977
|