When evaluating CT measurement data, extremely fast and precise automated algorithms are necessary in order to analyze and process the point clouds that are often very large. One important method in this context is the comparison of the CAD model with the measurement data, especially when it comes to gaining a quick overview of the overall component and its discrepancies from the ideal model. To do this, all the distances between the measurement data and the model have to be calculated. The calculated distances can then be color-coded and visualized on the model. Efficient algorithms have been developed at Fraunhofer IPA to calculate these discrepancies.
Another method for analyzing measurement data is the best-fit of geometric primitives, such as spheres or cylinders, and can be achieved using the best-fit method developed at Fraunhofer IPA with integrated segmentation. The method is not only extremely fast and precise but is also capable of analyzing measurement data without a CAD model. In this way, discrepancies in shape can be calculated as well as distances, angles and diameters.
With industrially manufactured parts, in particular cast parts, wall thicknesses often have to be verified. A measuring tool has been developed at Fraunhofer IPA to measure the thickness of the remaining material. The tool is highly versatile because measurements are made directly in the STL data. To allow industrial implementation, the algorithm has been optimized with regard to speed. Distances are visualized in color-coded form, thus enabling walls that are too thin to be easily identified. The advantage of this method over a simple comparison of CAD data with measured data is that the real wall thickness is measured.