Multi-Directional Illumination: A Novel 3D and BRDF Approach in Integrated Heritage Digitization

Abstract

In the proposed solution a unified data structure is acquired during measurement of a 3D object. It consists of geometrical data in the form of point clouds with additional spectral and angular reflectance information. High spatial correspondence can be achieved by the use of a single matrix detector to collect all measurement modalities. Resulting from previous works, the full processing path for multimodal measurement datasets of cultural heritage objects has been described in literature. We present the current advances in an integrated digitization approach by means of the described system with one additional step toward overall hardware integration. We introduce a technique requiring only a single illuminating component for both angular and spectral reflectance estimation, which brings us closer to a universal tool for multimodal digital documentation. We demonstrate 3D models created with multi-directional data merging algorithms which increase the overall quality of models based on high data density and partial redundancy. Two system configurations are presented and used for comparison of possible applications, yielding different measurement strategies in data acquisition process. Also described is a method of calibration for the directional illumination array along with a description of shadow mapping algorithms which can facilitate from this method.