Detection of shape deviations and measurement errors by a point cloud analysis

For the enhancement of technical workpiece surfaces with even larger dimensions, the application of microstructures on the surface is an appropriate way to improve the fitness for use without changing the properties of the basic material. Considering the extremely small dimensions of approximately 5–20 μm of the applied microstructure, the quality assurance faces new challenges related to the obtainment and evaluation of measurement data. This article presents an approach for the automated detection of shape deviations of a microstructure, as well as the detection of measurement errors during an optical or tactile measurement. The explained algorithm is based on the analysis of the measurement points within a point cloud by observing the distances between the single points. To illustrate the disturbance in the measurement point cloud every point is evaluated by an adaptive weighting function. The weighting of each measurement point can then be visualized by plotting the whole point cloud according to a corresponding color scale. The suitability of the point cloud analysis is demonstrated by the examples of a shape deviation (artificial groove) and a measurement error, occurred by measurement via confocal microscopy.