Spectral variations in HSI signatures of thin fabrics for detecting and tracking of pedestrians

Several hyperspectral imaging (HSI) systems have been developed with the intent of utilizing spectral diversity in a scene to characterize the background and identify targets of interest. With the increased spectral information, targets much smaller than the spatial extent of a pixel (e.g., pedestrians) can be detected and identified. However, several challenges arise when employing HSI systems for use in the detecting and tracking of pedestrians. Most notably, the signature for the target of interest may or may not be immediately available to the system. The successful detection of sub-pixel targets with any degree of accuracy relies on a priori knowledge of the spectral signature for the targets of interest. Previous work has shown that even if a spectral signature for a pedestrian was acquired at the time of detection, as the pedestrian moved through a cluttered environment its spectral signature changes. This work extends the understanding of the induced spectral variation in human spectral signatures in cluttered environments. The goal of this work was to isolate the spectral reflectivity of highly transmissive targets independent of the background. Using a doubling method, the effects of reflective backing materials on the signature of a thin fabric is presented. Also, an issue with tracking a pedestrian from full illumination into the shadow is considered. Reflectance factor signatures were measured using target reflectivity measured both in the lab and in the field to assess spectral variability and detectability.