高级检索

叶片破损区域边界的自动提取算法研究

Automatic Method for Broken Boundary Extraction

  • 摘要: 为了提高叶片的维修效率,提出一种叶片破损区域边界的自动提取算法.该算法主要包括两部分:区域分割和边界重构.在区域分割算法中,通过曲率估算、种子栅格选取和区域生长等步骤将破损区域的数据点从原始点云中分离出来;在边界重构算法中,通过多边形细分和离散光顺算法重构破损区域边界.应用实例表明,文中算法效率高、稳定性好,能够从散乱的点云数据中直接获取高质量的区域边界信息.

     

    Abstract: Turbine blades, working in high pressure and temperature environment, need repair regularly. To improve the efficiency of blade repair, an automatic method for broken boundary extraction is proposed in the paper. The extraction procedure consists of two steps: segmentation and boundary reconstruction. The segmentation extracts the damaged region from point cloud with the processes of curvature estimation, seed grid selection and region growing. This step can be completed in a short time through a 4D Shepard surface based curvature estimation method. In the second step, boundary reconstruction is approached by polygon subdivision and discrete fairing. A new global fairing indicator and a special given tolerance guarantee that boundary can be smoothed while preserving the required accuracy. The proposed method can perform without any user interaction, and it can also be used in other applications, such as the repair of molds and dies. Several examples are given to show its efficacy.

     

/

返回文章
返回