基于3D点云平均曲率估算的损伤参数计算  

Damage Parameter Calculation Based on 3D Point Cloud Mean Curvature Estimation

作　　者：张玉燕[1,2] 渠文涵 郭全丽[1,2] 任锐[3] ZHANG Yu-yan;QU Wen-han;GUO Quan-li;REN Rui(Institute of Electrical Engineering,Yanshan University,Qinhuangdao,Hebei 066004,China;Key Lab of Measurement Technology and Instrumentation of Hebei Province,Qinhuangdao,Hebei 066004;Academy of Armored Forces Engineering,Beijing 100072,China)

机构地区：[1]燕山大学电气工程学院,河北秦皇岛066004 [2]燕山大学测试计量技术及仪器河北省重点实验室,河北秦皇岛066004 [3]装甲兵工程学院控制工程系,北京100072

出　　处：《计量学报》2018年第5期609-614,共6页Acta Metrologica Sinica

基　　金：国家自然科学基金(61403333)

摘　　要：针对高速滑动电接触滑轨表面的微小损伤,基于激光扫描3D点云平均曲率估算的原理,提出一种表面微小损伤特征参数的定量表征方法。对3D点云数据进行去噪、滤波平滑、数据精简、k-邻域搜索等处理后,进行局部二次曲面拟合,估算出点云的平均曲率,设定曲率阈值实现损伤区域特征点云提取,对损伤区域进行定位,进而计算被测滑轨表面的微小损伤的体积及质量损失。实验研究表明:被测滑轨表面损伤质量的测量精度可达mg级,验证了该方法的有效性。Aiming at the surface damage measurement of high speed sliding electric contact rail, a method of evaluating damage mass based on mean curvature estimation of 3D laser point cloud is presented. The laser scanner is used to obtain point cloud data which will be denoised, filtered, reduced, and k neighborhood searched. Next step of data processing is local quadric surface fitting, then the mean curvature of the point cloud data is estimated. The curvature threshold is set to achieve the extraction of feature point cloud to achieve damage area. According to detection results, the damage volume and mass loss are calculated. The experimental results show that the measurement precision can reach mg level. Moreover, the proposed approach is feasible and effective to measure the surface damage.

关 键 词：计量学 滑动电接触 滑轨 点云数据 曲率估算 表面损伤 

分 类 号：TB92[一般工业技术—计量学]