|
White light scanning represents a significant milestone in the process of digitization. Not only is the subject untouched in the scanning process, but the quality and quantity of data acquired from the subject is extremely high. While conventional contact techniques may acquire up to a few thousand points in an hour, white light scanners can acquire up to 10 million points in the same amount of time. The benefits are obvious: more data in less time. With a high-density scan, every surface detail can be discerned; the thickness of a tape line, the veined texture of a maple leaf, even a fingerprint. Once acquired, scanned data is easily transformed into surface models that can be manipulated, scaled, animated, rendered, sectioned, or surfaced, and output as plots, stereolithography models, or CNC machine toolpaths.
The optical scanning process described: · Multiple scans are generated by freely positioning the scanner's "point of view" about the object. By moving the scanner independently of an absolute coordinate system, an operator is able to easily scan cavities and undercuts that would be inaccessible by other digitization tools. More importantly, the open envelope imposes no limitation on the size of the subject. · The multiple scans are aligned then merged together into a single point cloud by means of point assimilation (redundant points are thrown out). Surface continuity can be held within .0000001" · The merged XYZ point cloud is "skinned". The individual points are interconnected to form a continuous mesh surface comprised of thousands or millions of three-sided polygons. · The polygon mesh model can be output at multiple resolutions, to various platforms depending on the application - high resolution for small articles output to a CNC toolpath, or low resolution for use in animation. · A nurbs surface can be generated from the mesh data. This is especially valuable if a higher degree of surface quality is needed. New software enables us to generate selectively parameterized nurbs surface models directly from a polygon mesh, without having to perform the tedious task of interpreting cross sectional data (these models are virtually seamless and lack the faceted characteristics of polygonal data). Depending on the degree of detail and hard breaks in the surface, parameterization can vary from one surface feature to another.
Using a Renishaw probe attached to a Tarus 3 axis CMM machine, we are able to selectively digitize important surface characteristics of an article. For example, the pivot points of an articulated chair, the hyrdrodynamic profile of a boat hull, or the character lines defining the location of an integrated lens assembly, can be isolated into specific data sets, merged into existing surface files and used for engineering studies.
|