Optical Properties of Hygroscopic Materials for Background Matching

Sammanfattning

Background matching is a common camouflage strategy used by animals, where color and patterns are used to blend with the surrounding environment. Humans also utilize this principle in military applications to reduce the contrast between an object of interest and its background. Recent advances in sensor technology, particularly in the short-wave infrared (SWIR) (900–2500 nm) range, have introduced new requirements for camouflage materials intended for effective background matching. SWIR imaging is highly sensitive to water due to its distinct absorption features in this spectral range. Since natural backgrounds consisting of leaves and vegetation generally contain more water than conventional camouflage materials, a pronounced contrast arises between object and background. Controlling the water content in a material therefore becomes crucial for effective background matching in SWIR. Hygroscopic materials may enable such control due to their ability to absorb and retain water. Previous studies at the Swedish Defence Research Agency (FOI) in the SWIR range demonstrated that hydrated textiles exhibit lower reflectance than dry textiles, with woven cotton and polyester showing reflectance levels similar to foliage. However, the quantitative relationship between water content and optical properties in SWIR has not previously been investigated. Addressing this knowledge gap is the main objective of this thesis, which aims to develop a novel methodology for quantifying the relationship between water content in hygroscopic materials and their sensor response in a SWIR imaging system (µy in ROI [DN]), and to use this relationship to reduce contrast against selected leaves. Several hygroscopic materials were investigated using a novel experimental setup consisting of a SWIR camera, analytical scale, fan, and halogen lamp. The obtained data, including water content [g/cm3] and µy in ROI [DN], were analyzed, normalized, and statistically modeled to identify suitable relationships between water content and the sensor response in a SWIR imaging system. The developed models were subsequently applied to SWIR imaged leaves to estimate water content intervals required for reducing contrast between the materials and the leaves in SWIR. Validation was then performed by applying the predicted water content to the materials and evaluating whether the target µy [DN] of a leaf could be achieved. The results yielded three non-linear declining models, all showing that increasing water content reduces µy [DN] in SWIR due to enhanced light absorption by water. The most promising model for background matching applications was based on raw image data (µy [DN]) and water content [g/cm3] data from non-woven polyester and non-woven viscose/polypropylene textiles. The predicted water content interval required in these materials to reduce contrast against leaves ranged from 0.3 to 1.2 g/cm3. Validation demonstrated that the model performs well for the materials on which it was developed, particularly for non-woven polyester, but does not generalize to structurally different materials due to differences in internal structure, water distribution, and optical behavior. Overall, the findings indicate that the developed model can be used to predict the water content [g/cm3] required in non-woven polyester and non-woven viscose/PP textiles to match the µy [DN] of a selected leaf in the SWIR range. The study provides a solid foundation for the future development of smart background matching solutions in SWIR.