顾及强度信息和地形信息的南方丘陵区机载LiDAR数据滤波研究  

Research on Airborne LiDAR Data Filtering Considering Intensity Information and Terrain Information

作　　者：陈光洲 岳东杰[2] 王亚芹[1] Chen Guangzhou;Yue Dongjie;Wang Yaqin(Department of Water Conservancy Engineering,Zhejiang Tongji Vocational Collegeof Science and Technology,Hangzhou 311231,China;School of Earth Sciences and Engineering,Hohai University,Nanjing 211100,China)

机构地区：[1]浙江同济科技职业学院水利工程系,浙江杭州311231 [2]河海大学地球科学与工程学院,江苏南京211100

出　　处：《甘肃科学学报》2021年第3期55-60,共6页Journal of Gansu Sciences

基　　金：浙江水利科技计划项目(RC1828)。

摘　　要：针对南方丘陵区地形起伏较大的特点,结合点云的强度信息,研究适合于该区域的机载LiDAR点云滤波算法。首先,对强度信息进行改进的K-MEANS聚类,并根据距离评价函数最小准则寻找最优聚类数。其次,结合渐进三角网滤波算法能较好顾及地形起伏的优点,对不同类型的点云设置不同的构网优先级,得到不同的判据阈值,以此进行渐进三角网滤波。选取南方丘陵区浙江省某地两处的机载LiDAR数据进行实例验证,并与TerraScan的滤波效果进行比较。结果表明,新算法精度高于TerraScan滤波,其滤波效果更优,具有一定的可靠性。In view of the features of large topographic relief in hilly areas of southern China,this paper studies on airborne LIDAR point cloud filtering algorithm suitable for this region combining the intensity information of point cloud.First,intensity information is clustered with improved K-MEANS clustering algorithm.The best clustering number is chosen according to the minimum criterion of distance evaluation function.The progressive triangulation filtering algorithm can take into account the advantages of terrain undulation.Different priority of network construction is set for different types of point cloud,and different threshold of criterion is obtained,so as to filter the progressive triangle network.Two cases study of airborne LiDAR data in a place of Zhejiang Province in southern hilly region is tested,which results are compared with TerraScan filtering.The results show that the accuracy of the FCIT algorithm is higher than that of TerraScan filter and has a better filtering effect.It has a certain degree of reliability.

关 键 词：机载LIDAR 南方丘陵区 点云滤波 强度信息 地形信息 

分 类 号：P23[天文地球—摄影测量与遥感]