Comparison of inverse estimation methods of radiative reflection properties using temporal and directional IR radiances

Abstract

It is essential to prepare reliable material and radiative surface properties in predicting the accurate infrared signature of objects on a scene. Especially it is difficult to procure the radiative reflection property expressed by the Bidirectional reflectance distribution function (BRDF) due to the limited information in the literature. Recently, studies on inverse estimation of the radiative properties based on experimental data are conducted to obtain radiative reflective properties required for the infrared image generation. In this study we compare the two different inversely estimated radiative reflective properties obtained respectively from the daily measured radiance data with a fixed angle of measurement and the measured directional radiance data at a fixed time in a lab environment. The radiative properties in Mid-wavelength infrared (MWIR) region of the painted steel are estimated by finding the best fitted model parameters for the Sandford-Robertson BRDF model using the Repulsive particle swarm optimization (RPSO) algorithm by using those two different radiance data. The estimated radiative reflective properties obtained from the two different radiance data show fairly good agreements with the reference data but the measured directional radiance data at a fixed time show better results. The directional radiance data measured at a fixed time in a lab environment could be a better way of obtaining the radiative surface properties of newly developed paints when an appropriate measuring device for the directional radiance is in hand.