3D Data Fusion
 The present 3D model of Tiwanaku includes multi-scale, multi-temporal, and multi-source "organized” point-clouds co-existing within a single coordinate system, and containing hundreds of millions of points. A point cloud is a dense collection of three-dimensional observations (in the form of points) describing the topology and morphology of a site. Each point in the cloud may also contain attributes that describe additional properties of the surface (e.g. color as measured by digital cameras, reflectivity of infrared light from the LIDAR source, resistivity and magnetic susceptibility of the soil). An “organized” point cloud contains additional information in the form of connectivity among points in the cloud. For example, points in the Tiwanaku site model “know” which points they are closest to and whether they belong to the same surface as those points. This organizing information facilitates efficient processing of the huge dataset, but more importantly, provides guidance in automated feature extraction, analysis and model query.
The site model format allows for the introduction of a wide range of data, including three-dimensional terrestrial and airborne LIDAR scans, subsurface geophysics observations and aerial and terrestrial images. Already, instruments and processing methodologies have been contributed to the Tiwanaku site model including terrestrial LIDAR scans, photogrammetric surface reconstructions using historical aerial photography, total station survey and a variety of geophysical instruments. Observations from other instruments, such as modern digital camera, digitized film from historical photography or airborne LIDAR, may be also be readily assimilated into this model.
|
