Global shape defect detection for 3D printed medical plates using medial skeletons and multiresolution hierarchies

초록

3D printing is commonly used for producing patient-specific plates. Shape defects can arise from the printing environment or from the heat treatment. A common method for detecting and analyzing shape defects involves measuring the minimum distance between corresponding surface points on the CAD model and the part. Global defects can cause large deformations in which the corresponding points are far apart. Prior work does not address this case. We present an algorithm that detects global shape defects by modeling them as deformations of CAD models. We acquire a 3D model of the printed part, deform this model to the original CAD model using medial skeletons and multiresolution hierarchies, compute point correspondences between the skeletons, and map them to corresponding points on the original models. Through experiments, we show that this algorithm detects global shape defects effectively and supports accurate defect analysis.

키워드