An infrared DoLP calculation model with parameters instantiation for high-temp rough surfaces

Abstract

Infrared polarimetric imaging has gained more attention in recent years in various defense applications. As the temperature of an object rises, the infrared polarization on its surface becomes stronger in room-temp environments. This polarization mainly depends on the thermal emission and is slightly affected by the reflection of ambient radiation. Due to the difficulty in solving the relevant sample parameters in the infrared band, the complexity of infrared polarization modeling for high-temp surfaces exceeds that of visible and near-infrared. Based on the polarized bidirectional reflectance distribution function (pBRDF) and Kirchhoff’s law, this work first derives the emissivity vector. Then, a hybrid infrared degree of linear polarization (DoLP) calculation model framework for high-temp rough surfaces is proposed. This framework consists of thermal emission and the reflection of ambient radiation. The reflection model component is simplified by uniformizing the ambient incidence radiation on surfaces. An infrared DoLP measurement experiment is conducted on three sample surfaces at multiple high-tempe points. Considering the insufficiency of measurement data at a single high-temp point, model parameters are inverted using data from two high-temp points. The genetic algorithm is selected as the inversion method, and an instantiated hybrid model is obtained based on the optimal solution of parameters. At multiple high-temp points, the DoLP curve predicted by the hybrid model is consistent with the measured curve. When the observation angle is less than $75°$, the absolute residual between the predicted and measured DoLP is less than 0.005. The experimental results demonstrate that the proposed hybrid infrared DoLP calculation model can be applied to predict and simulate the polarization characteristics of high-temp objects in room-temp environments. Moreover, the proposed inversion approach can accurately estimate the material surface parameters.