江西科学2016，Vol.34Issue(6)：875-878,4.DOI:10.13990/j.issn1001-3679.2016.06.032
K-邻域法对不同形态植被点云数据特征的提取
Feature Point Extraction of Irregular Natural Surface Features of Point Cloud Data Based on K-Nearest Neighbors
程亚丽 1李向新 1李鹏飞1
作者信息
- 1. 昆明理工大学国土资源工程学院，650093，昆明
- 折叠
摘要
Abstract
By using the 3D laser scanner to get point cloud data as data sources,and through the classification of cutting to get point cloud data of different natural features. K-neighborhood tree method is used for irregular natural features of point cloud data to neighborhood search,which by set-ting the value of the scale parameter k to the height or low of different on the extraction of feature points,and then compared the result of the feature points. The result show that scale parameter set-tings have significant impact on the larger density of low vegetation point cloud,but have week obvi-ous effect on the low density of high vegetation. Setting the scale parameter between 3 to 5,the exper-imental effect is remarkable.关键词
K-邻域法/点云数据/不同形态植被/尺度参数/特征提取Key words
K-nearst neighbors/point cloud data/different forms of vegetation/scale parameter/fea-ture extraction分类
信息技术与安全科学引用本文复制引用
程亚丽,李向新,李鹏飞..K-邻域法对不同形态植被点云数据特征的提取[J].江西科学,2016,34(6):875-878,4.
