Surface Reconstruction and Mesh Generation Using Reverse Engineering Approach

Geometry modeling and mesh generation are pre-processing steps in computational engineering as many computational engineering disciplines require a valid numerical mesh for solving the respective governing equations. However, mesh generation is often the most labor-intensive phase during the entire numerical simulation cycle, as a user needs to cautiously maintain the fidelity of the geometry while creating high quality meshes on and about the geometry. Non-Uniform Rational B-Spline (NURBS) is currently the most robust and most popular technology that is currently used as the de facto geometry representation between CAD and mesh generation systems. Unfortunately, not all the geometry acquired by the user can be readily represented in NURBS. The focus of this paper is to describe the approaches that address issues associated with structured mesh generation in the process of reverse engineering. These approaches utilize 3-D laser scanners or contact scanners to digitize the 3-D models and obtain point cloud that defines the geometry. These points are used to generate high quality 3D structured surface mesh through an interactive process.