Research And Implement Of Surface Feature Extraction Based On Line Element Geometry

Because of the advantages on accuracy and expeditiousness feature-based surface reconstruction becomes a hot topic in reverse engineering. Especially, 3D shape recognition and feature extraction is the difficulty and key point of the feature-based reverse engineering.Based on the review and analysis of existing research achievements of feature technology, the thesis presents a method for the recognition and feature extracting of surfaces from 3D data. Especially studied the segmentation method based on feature classification and feature approximation algorithms based on the segmentation results. The main research and achievement as follows:1. The thesis presents a method for 3-D shape segmentation based on line element geometry and linear complex. It is the preparation of feature extraction of single-feature analytic surface. Line element geometry is employed to effectively perform the presentation of complex objects according to surface type. K-Local-RANSAC algorithm use linear complex to perform the classification. The discrete region growing mode improves the efficiency obviously.2. An automatic surface type understanding and feature extraction method was designed. The author uses equiform kinematic surface in combination with PCA to recognize the type of surface and approximate feature parameters. Besides an axis adjust algorithm and a generatrix predigest algorithm was presented in order to overcome the disadvantage of iterative methods and improves the result on the aspect of accuracy.All of the algorithms presented above have been implemented in a prototypical system and tested with data obtained from objects in real world. Results of these examples indicate the correctness of the algorithms in this thesis.