Image-based COTS Bidirectional Reflectance Distribution Function (BRDF) Measurement

TECH FOCUS AREAS: Directed Energy TECHNOLOGY AREAS: Space Platform; Materials OBJECTIVE: Develop an image-based system operated by a minimally trained operator for rapid BRDF generation. DESCRIPTION: Generating realistic and radiometrically accurate images of modeled scenes, both for computer graphics and remote sensing applications, requires a complete description of the reflection from all material surfaces in the scene at all incidents and reflected angles and all wavelengths of interest. This description of the reflective behavior is commonly referred to as the BRDF. Many mathematical models describing the BRDF exist, some better suited to certain classes of materials than others. However, they all contain material-specific parameters that are not readily available and in the best case are derived from measured BRDF data. Instruments to make BRDF measurements are generally highly specialized, purpose-built, rare, and expensive. The measurements are also time-intensive and thus usually sparsely sampled. These drawbacks have been successfully addressed by previous efforts to generate BRDFs using an image-based technique.1 These efforts utilized multiple cameras to illuminate a curved sample at different angles and capture many reflected angles simultaneously in each image. Computer vision and image post processing eliminates the need for precise positioning and generates millions of separate reflection measurements that cover the whole reflection hemisphere. However, the imaged-based system previously developed was a proof-of-concept instrument requiring expert operation and image processing. For a space domain awareness utility, this methodology has not been integrated into a commercially available system that can be operated by a minimally trained operator for rapid BRDF generation. Also, in the years since, enabling technologies have developed to further decrease the barrier to entry to performing these measurements, including cell phone cameras, 3D printing, and app-based software. Decreasing equipment cost and increasing user-friendliness would enable image and computer vision based BRDF measurement of individual materials over representative samples, on-site measurements, and more commonplace use of measurements over BRDF estimates. BRDF are generally used for: Material identification of spacecraft surfaces for space domain awareness Characterization and identification for RSO (Resident Space Objects), aircraft, and missiles Understanding lighting conditions in an image due to foreign light sources (albedo, planetshine, retroreflection, ringshine, etc) Modeling visible glints in space and aerospace applications Rendering for computer applications and video processing PHASE I: An initial solution to this topic would include the basic hardware required to re-create the image based BRDF generation demonstrated by Marschner, as detailed in reference 1. This would include the sample and camera mounts and COTS cameras to capture the necessary BRDF images. However, the major effort is in developing the software to enable BRDF generation. This includes computer vision algorithms to automatically generate the relative angles in the scene from the images themselves, combination/averaging of image pixels representing the range of BRDF angles, and fitting of the data to several popular BRDF models. Emphasis in the development is to be placed both on precision of the measurement and on reducing the investment necessary in the system. This could range from a bare-bones approach utilizing 3D printed parts and cell phone cameras, to specialized mounts and photographer-grade digital cameras. PHASE II: Phase II efforts would seek to decrease the total cost of the system and the skill level and input required by the user. This may include transitioning from professional-grade cameras to consumer-grade or cell phone cameras, deriving more of the necessary inputs from the images themselves, rather than user inputs, or developing analysis software that self-guides the user through the BRDF generation process. PHASE III DUAL USE APPLICATIONS: Phase III efforts would create a mass market product utilizing widely available hardware (ideally cell phone cameras) and packages the analysis software into a cell-phone application. This would allow BRDFs of individual materials or objects (rather than representative articles) to be captured on site and used to populate models for remote sensing or imported into individualized computer-generated scenes. REFERENCES: 1. Stephen R. Marschner, Stephen H. Westin, Eric P. F. Lafortune, and Kenneth E. Torrance, "Image-based bidirectional reflectance distribution function measurement," Appl. Opt. 39, 2592-2600 (2000) KEYWORDS: Modeling and Simulation; Remote Sensing; Computer Vision; Space Domain Awareness; Image Rendering/processing