点云模型特征的提取算法  

Algorithm for Extracting Sharp Features from Point Cloud Models

作　　者：邹冬[1] 庞明勇[1] 

机构地区：[1]南京师范大学教育技术系,南京210097

出　　处：《农业机械学报》2011年第11期222-227,共6页Transactions of the Chinese Society for Agricultural Machinery

基　　金：国家自然科学基金资助项目(60873175);安徽省高校省级自然科学研究资助项目(KJ2010B423;KJ2010B142)

摘　　要：提出一种基于移动最小二乘法的点云模型尖锐特征提取算法。首先使用投影残差来识别潜在的特征点,然后采用一种优化的主元分析法光顺潜在的特征点,再利用改进的折线生长方法生成特征线,最后为模型建立角点完善提取的特征线。实验表明,本文算法运行稳定,性能优于其他算法,可以准确地捕捉点云模型上的特征线。Based on the moving least squares method,a method for extracting feature curves from point clouds was presented.Firstly the algorithm calculated projection residuals and potential feature points were identified in point cloud model.The potential feature points were then smoothed by employing a modified version of the principal component analysis approach.Subsequently,a feature-polyline propagation technique was used to approximate the feature points by a set of polylines.Finally the feature curves were optimized by the algorithm to resolve gaps and recover the junctions.Experiments show that the algorithm is very robust,and it can extract feature curves from various point clouds.

关 键 词：点云模型 特征提取 移动最小二乘法 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]