Generatrix surface extraction for hollow turbine blade reverse engineering based on CT image sequence

To investigate and redesign turbine blades in reverse engineering, 3-D geometric models need to be built by reconstruction. However, the process of acquiring point cloud data for reconstruction is made difficult because of the complexity of the hollow turbine blade geometry In particular, the existence of partitions, disturbance columns and trailing edge holes designed for air cooling and turbulent makes the segmentation of generatrix surfaces and local geometric features a challenging sur task. Focusing on this problem, the paper presents a novel approach to extract generatrix point cloud of a hollow turbine blade from its CT image sequence for later 3-D geometric feature reconstruction. This approach increases surface point location accuracy by employing a subvoxel edge detection method and effectively extracts generatrix surface points by employing a guiding strategy and B-spline fitting. Finally, experiments and results are presented to validate the effectiveness of this approach for hollow turbine blades.