Empirical broadband polarimetric bidirectional reflectance distribution function measurement framework for application to remote sensing data

The bidirectional reflectance distribution function (BRDF) is essential to remote sensing, computer graphics, and material science applications where it aids in the development of target detection and recognition algorithms, photo-realistic rendering models, and atmospheric characterization tasks. BRDF measurements are time-intensive and may require dense, full hemispherical sampling that often results in millions of individual measurements on a given material surface conducted at precise sensor and illumination source geometries. Measurements are typically obtained at a discrete set of wavelengths and fit to physics-based models to estimate material parameters of interest. Although useful for material characterization tasks and phenomenology studies, applying these models effectively to remote sensing data is often nontrivial. Researchers have alternatively explored empirical approaches using three-dimensional targets in conjunction with broadband illumination sources and image-based sensors to parallelize the measurement process for constructing parameterized material measurements. Such empirical databases can be used as lookup tables and are often more conducive for applying to image data collected with similar broadband sensors. In this work, we adapt such an outdoor image–based broadband polarimetric BRDF (pBRDF) measurement strategy previously proposed by our group for use in a laboratory environment. We describe the adaptations employed to allow for equivalent empirical pBRDF measurements to be conducted in a laboratory environment and demonstrate that the measurements validate well against their corresponding outdoor measurements.