Mid-wave infrared target source characteristics for focal plane applications

The U.S. Army is currently investigating the differences between various bands in the midwave and long wave infrared spectrum. A holistic approach to quantifying scene information is used in previous research. That is, both natural backgrounds and vehicles are present in scenes when correlation analyses are performed. Similar research has also been performed using hyperspectral imagers. Hyperspectral imagers inherently have poor signal-to-noise ratio (SNR). In this research, a mid-wave infrared broadband sensor was cold-filtered to provide four sub-bands in the mid wave region. A multi-waveband sensor was used to collect midwave infrared imagery of military vehicles and natural backgrounds. Three blackbody sources were placed at the same range as the vehicles for radiometric calibration. The goals were to collect radiometrically corrected data of various targets and process this data for comparative analysis. The images were segmented to remove all unwanted imagery from the images under observation. Correlations were performed to assess the differences in information content.