|
Abstract:
|
With the development of 3D optical scanning technologies in Industry 4.0, 3D surface topography measurement plays an increasingly important role in many fields such as product quality inspection in additive manufacturing, gauge capability analysis, and firearm identification in forensic science. It is necessary to evaluate the similarity among surface topography measurements in practice. In additive manufacturing, if two surface topography measurements are portraying the same surface, we define them as matched pairs; otherwise, they are non-matched pairs. Yet due to the complexity of surface topography measurements, it is challenging to measure the similarity of measurements and identify matched or non-matched pairs. In this paper, we establish a thorough framework of 3D surface topography measurements comparison. The proposed framework generates a scaled similarity score for a pair of surface topography measurements and provides two decision-making methods to distinguish matched and non-matched pairs. We illustrate the framework and evaluate the classification performance on data measured from additive manufactured samples in a large scale.
|
