Processing of 3D Unstructured Measurement Data for Reverse Engineering

Abstract

One of the most difficult problems in reverse engineering is the processing of unstructured data. NURBS (Non-uniform Rational B-splines) surfaces are a popular tool for surface modeling. However, they cannot be directly created from unstructured data, as they are defined on a four-sided domain with explicit parametric directions. Therefore, in reverse engineering, it is necessary to process unstructured data into structured data which enables the creation of NURBS surfaces. This paper presents a methodology to processing unstructured data into the structured data for creating NURBS surfaces. A projection based method is established for constructing 3D triangulation from unstructured data. An optimization method is also established to optimize the 3D triangulation to ensure that the resulted NURBS surfaces have a better form. A triangular surface interpolation method is established for constructing triangular surfaces from the triangulation. This method creates five-degree triangular surfaces with C1 continuity. A series of segment data are obtained by cutting the triangular surfaces with a series of parallel planes. Finally, the structured data is obtained by deleting repetitive data points in each segment data. Results demonstrate the efficacy of the proposed methodology.