An increasing reliance on infrared (IR) sensors for accurate detection, classification, and tracking of Time-Critical Targets (TCTs) in background clutter has resulted in a growing need for physical model-based, yet real-time and affordable, weather-dependent diurnal IR image simulation of TCTs embedded in geospecific backgrounds for weapons systems training, mission planning, mission rehearsal as well as weapon system development and testing. The problem is that model-based IR image simulation requires time-consuming estimation of IR model parameters for a large number of objects, materials, and geographic areas. The problem becomes especially severe when simulating imagery of denied access areas where IR characteristics of background materials and objects for every pixel on the ground are unavailable or difficult to obtain.

Therefore, we have developed an approach to weather-dependent diurnal IR background image simulation based on IR model parameter estimation from Multi-Spectral Imagery (MSI) such as available from commercial satellite and tactical reconnaissance sources. This paper describes the processes and results of thermal mass and visible emissivity estimation from Landsat Thematic Mapper visible and thermal band MSI data. Simulated thermal images of geospecific backgrounds at various times of day are shown using the estimated parameters and real geospecific weather data as inputs to AIRSIM - the US Air Force IR Synthetic Image Model. The significance of this approach is that the thermal models used can be applied equally well to targets and backgrounds. Use of the same physics-based model for both assures that there will be no false target/background contrast due to use of different models for each. Another major feature is that this approach allows automated generation of IR databases over large areas of terrain while retaining major thermal properties absent from DMA data, such as thermal mass. This allows for more complete and accurate representation of diurnal and weather effects in simulated imagery while achieving high realism from geospecific weather and terrain data.