Automatic data segmentation for geometric feature extraction from unorganized 3-D coordinate points

Citation
Jb. Huang et Ch. Menq, Automatic data segmentation for geometric feature extraction from unorganized 3-D coordinate points, IEEE ROBOT, 17(3), 2001, pp. 268-279
Citations number
18
Categorie Soggetti
AI Robotics and Automatic Control
Journal title
IEEE TRANSACTIONS ON ROBOTICS AND AUTOMATION
ISSN journal
1042296X → ACNP
Volume
17
Issue
3
Year of publication
2001
Pages
268 - 279
Database
ISI
SICI code
1042-296X(200106)17:3<268:ADSFGF>2.0.ZU;2-0
Abstract
In this paper, a systematic approach is proposed to automatically extract g eometric surface features from a point cloud composed of a set of unorganiz ed three-dimensional (3-D) coordinate points by data segmentation. The poin t cloud is sampled from the boundary surface of a mechanical component with arbitrary shape. The proposed approach is composed of three steps. In the first step, a mesh surface domain is reconstructed to establish explicit to pological relation among the discrete points. The topological adjacency is further optimized to recover the second order object geometry. In the secon d step, curvature-based border detection is applied on the irregular mesh t o extract both sharp borders with tangent discontinuity and smooth borders with curvature discontinuity. Finally, the mesh patches separated by the ex tracted borders are grouped together in the third step. For objects with co mplex shape, multilevel segmentation scheme is proposed for better results. The capability of the proposed approach is demonstrated using various poin t clouds having distinct characteristics. Integrated with state of art scan ning devices, the developed segmentation scheme can support reverse enginee ring of high precision mechanical components. It has potential applications in a whole spectrum of engineering problems with a major impact on rapid d esign and prototyping, shape analysis, and virtual reality.