摘要
地面三维激光点云数据的植物器官分割,是林业信息化测量中的基础性工作之一。本文在点云数据颜色相近、结构复杂的情况下,首先提出了一种新的局部切平面分布特征,并构造了融合原始扫描数据、散点空间分布特征、法向分布特征的多维融合特征,能够更为有效地表征不同类别的植物器官。其次在分类器选择上,采用标准SVM,PSVM,GEPSVM三种分类器作为对比,后续使用图割理论进行再分类,加强分类效果。根据多种比较实验表明,本文提出的多特征融合分割方法能有效对植物器官的点云数据进行分类,其识别率可达到98%以上。
The segmentation of foliage organs from 3Dpoint clouds is an elemental work of forestry informatization measurement.However,the foliage point cloud data has a similar color,and the point construction is complex which can not be expressed easily.Therefore,a novel feature called local tangent plane distribution is proposed,and fused with original data,scatter spatial distribution and normal distribution to construct a multi-dimension feature,which can characterize different foliage organs more effectively.Then three kinds of classifiers,including standard SVM,PSVM,GEPSVM,are used as a comparison.And then the graph cut is also utilized for a re-classification at subsequent processing to improve the classification performance.A variety of comparative experimental results show that the proposed mutli-dimension feature segmentation method can effectively classify the foliage organs from point cloud data.The recognition rate can reach 98%.
出处
《数据采集与处理》
CSCD
北大核心
2015年第5期1054-1061,共8页
Journal of Data Acquisition and Processing
基金
国家自然科学基金(31300472)资助项目
江苏省自然科学基金(BK2012815
BK2012418)资助项目
关键词
点云
植物器官
数据分割
多维特征
支持向量机
point clouds
foliage organs
data segmentation
multi-dimensions feature
support vector machine
