Accurate Segmentation Technology of Point Cloud Data Based on Feature Template
LIANG Biao 1 , LI Shuanggao 1 , HUANG Xiang 1 , LI Gen 2 , PENG Yun 1
1. College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China;
2. Suzhou Research Institute of Nanjing University of Aeronautics and Astronautics, Suzhou 215000, China
Abstract:With the continuous development of digital measurement technology, the scale of acquired three-dimensional point cloud data has increased dramatically, and the segmentation technology of point cloud data has become particularly important in data processing. In this paper, combining the characteristics of the measured point cloud data of aircraft parts and components, it proposes a precise segmentation technology of point cloud data based on feature templates. First, extract the surface features to be recognized in the theoretical digital model, construct a feature template and store it in the template database; Then use the spatial grid method to establish the point cloud topological relationship and complete the rough segmentation of the point cloud based on the feature template; The point cloud normal vector is estimated by principal component analysis, and the surface-based adaptive region growing algorithm is used to further finely segment the coarsely segmented point cloud. Example verification shows that this technical method can achieve the purpose of accurate segmentation of point cloud data.
梁标,李泷杲,黄翔,李根,彭云. 基于特征模板的点云数据精确分割技术[J]. 航空制造技术, 2022, 65(11): 112-119.
LIANG Biao,LI Shuanggao,HUANG Xiang,LI Gen,PENG Yun. Accurate Segmentation Technology of Point Cloud Data Based on Feature Template[J]. Aeronautical Manufacturing Technology, 2022, 65(11): 112-119.