| Along with the development of ICT technology and introduction of feature technology into reverse engineering, the feature-based reverse engineering of ICT slicing image is proposed and has been becoming a hot topic of current research in this area. In this dissertation, some key technologies related to the geometric feature model reconstruction based on Industrial CT slicing dates are studied.It mainly includes:Image processing of ICT gray images is firstly accomplished. On the basis of analyzing the features of industrial CT images and the methods of existent image segmentation, the existing Otsu method is researched. The image is segmented through improved method of chaotic PSO and two-dimensional Otsu. The method has strong anti-noise ability and good segmentation results. At the same time, chaotic particle optimization is introduced and provides a new way for two-dimensional Otsu method of real-time applications.Classification and Presentation of features is analyzed and summarized thoroughly in variation design based CAD modeling methodology in reverse engineering. Curvature based feature identification method are synthesized. Theory of multi scale is introduced into section feature curve segmentation. A technical idea of automatic feature segmentation is proposed based on curvature multi-scale analysis. Based on curvature scale space, the overall contour geometric feature and the details feature are retained. A new technical approach is provided for capturing high level product structure.Based on the foregoing research, this paper studied the contour segmentation based on feature points, these feature points are regarded as the gist to recognize section feature. When the section feature is recognized, they should be fitted with LSM. When the section curve feature is fitted, the constrained optimization technique between the contour curves is studied.The research results of this paper can be an important reference for the future research. |
PDF Full Text Request