基于CSF的无人机影像数据点云滤波算法实现与应用  

Implementation and Application of Point Cloud Filtering Algorithm for UAV Image Data Based on CSF

作　　者：闵启忠 MIN Qizhong(Chongqing Survey Institute,Chongqing 401121,China)

机构地区：[1]重庆市勘测院,重庆401121

出　　处：《北京测绘》2021年第6期707-711,共5页Beijing Surveying and Mapping

摘　　要：基于无人机影像匹配点云数据技术,提出利用布料模拟滤波算法(Cloth Simnlation Filter,CSF)进行点云滤波计算,结果表明:布料模拟滤波算法在点云滤波匹配过程中,仅需要对六个参数进行简单设定就可以达到预期的地面点和非地面点的滤波匹配效果,地面点的匹配误差率为4. 9%,非地面点的误差匹配率为8. 4%,总体误差匹配率为6. 49%,表明布料模拟滤波算法匹配效果较好,匹配操作过程简单,具有较强的可操作性和可行性。Based on UAV image matching point cloud data technology,The cloth simulation filtering algorithm(CSF)was proposed to carry out the point cloud filtering.The results showed that cloth simulation filtering algorithm could achieve the expected filtering matching effect of ground points and non ground points only by setting six parameters in the process of Point Cloud Filtering matching.The matching error rate of ground points is 4.9%,and that of non ground points is 4.9%.While,the error matching rate of ground points is 8.4%,and the overall error matching rate is 6.49%.It showed that the cloth simulation filtering algorithm had better matching effect,simpler matching operation process,stronger operability and feasibility.

关 键 词：无人机(UAV)影像 点云匹配 布料模拟滤波算法 参数 误差率 

分 类 号：P208[天文地球—地图制图学与地理信息工程]