2016年
AUTOMATIC SEGMENTATION AND FEATURE IDENTIFICATION OF LASER SCANNING POINT CLOUD DATA FOR REVERSE ENGINEERING
2016 INTERNATIONAL SYMPOSIUM ON FLEXIBLE AUTOMATION (ISFA)
- ,
- ,
- ,
- 開始ページ
- 278
- 終了ページ
- 285
- 記述言語
- 英語
- 掲載種別
- 研究論文(国際会議プロシーディングス)
- 出版者・発行元
- IEEE
This paper describes the system for automatic segmentation and feature identification of unstructured point cloud data of laser scanning for reverse engineering application. The objective of this working is to assure that type of each extracted feature can be identified and classified automatically. The geometric forms such as a plane, cylinder, point, line, rule-surfaces, and free-form shapes are automatically extracted and identified from a segmented region of the point cloud. The identified features are then utilized for reverse engineering application in the registration process and reconstruction 3D CAD model.
The proposed method consists of segmentation process using region growing based on normal vector and curvature, feature extraction and identification of each using the geometry fitting criteria and utilizing the features into the registration process.
Through applying to a real case, we demonstrate that our proposed method is effective in segmentation and feature identification and applicable for reverse engineering.
The proposed method consists of segmentation process using region growing based on normal vector and curvature, feature extraction and identification of each using the geometry fitting criteria and utilizing the features into the registration process.
Through applying to a real case, we demonstrate that our proposed method is effective in segmentation and feature identification and applicable for reverse engineering.
- リンク情報
- ID情報
-
- Web of Science ID : WOS:000391854000050