Polygonal models are the most common representation of 3D data in computer graphics and in reverse engineering (RE) applications. The post-processing phase of such data (also known as mesh) to generate a CAD model is performed through several operations, such as polygon editing (mesh decimation, refining, smoothing, de-featuring, filling), curve extraction and surface fitting. These computer tools are mainly available in specific software for reverse engineering. In this paper a CAD-based methodology is proposed for planar and cylindrical features recognition starting from polygonal meshes. Applying segmentation algorithms and CAD methodologies, the mesh is converted into a featured CAD model. The proposed methodology is based on the analysis of the discrete curvature and presents two main steps. In the first step, similar regions are classified by means of a clustering algorithm (K-means) and then a growing mechanism is applied to create regions with similar discrete curvature. In the second one, regions are classified by means of their mean and gaussian discrete curvature. Then, plane and cylinder best fitting routines are applied in order to convert the polygonal model into CAD features. The proposed methodology has been implemented in the Rhino3D CAD system as Visual Basic (VB) interface, through the VBS language. A simple user-friendly GUI guides the user to import a polygonal mesh from which, by using the feature recognition algorithm, a featured CAD model is generated.

A CAD-based methodology for planar and cylindrical feature recognition

GERBINO, Salvatore
2008

Abstract

Polygonal models are the most common representation of 3D data in computer graphics and in reverse engineering (RE) applications. The post-processing phase of such data (also known as mesh) to generate a CAD model is performed through several operations, such as polygon editing (mesh decimation, refining, smoothing, de-featuring, filling), curve extraction and surface fitting. These computer tools are mainly available in specific software for reverse engineering. In this paper a CAD-based methodology is proposed for planar and cylindrical features recognition starting from polygonal meshes. Applying segmentation algorithms and CAD methodologies, the mesh is converted into a featured CAD model. The proposed methodology is based on the analysis of the discrete curvature and presents two main steps. In the first step, similar regions are classified by means of a clustering algorithm (K-means) and then a growing mechanism is applied to create regions with similar discrete curvature. In the second one, regions are classified by means of their mean and gaussian discrete curvature. Then, plane and cylinder best fitting routines are applied in order to convert the polygonal model into CAD features. The proposed methodology has been implemented in the Rhino3D CAD system as Visual Basic (VB) interface, through the VBS language. A simple user-friendly GUI guides the user to import a polygonal mesh from which, by using the feature recognition algorithm, a featured CAD model is generated.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11695/16453
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact