基于不规则三角网的LiDAR数据的边缘检测新算法  

A Novel Algorithm Based on Triangulated Irregular Network for Edge Detection from LiDAR Data

作　　者：宣贺君 苗启广[1] 刘如意[1] 郭雪[1] 

机构地区：[1]西安电子科技大学计算机学院,陕西西安710071

出　　处：《光学学报》2014年第12期307-313,共7页Acta Optica Sinica

基　　金：国家自然科学基金(61072109;61272280;41271447;61272195);教育部新世纪优秀人才支持计划(NCET-12-0919);中央高校基本科研业务费专项资金(K5051203020;K5051203001;K5051303018)

摘　　要：给出了一种三角形形变量的定义,并提出了基于不规则三角网(TIN)的LiDAR数据边缘检测新方法。将点LiDAR数据进行三角剖分,生成不规则三角网,计算TIN中每个三角形的形变量,根据三角形形变量的不同来确定处于地物目标边缘的三角形,对这些边缘三角形进行处理得到边缘点。针对LiDAR数据中可能由于河流等导致的数据空白区域,仅利用三角形形变量无法检测到所有边缘点的问题,提出了顶点到重心距离的平方和作为测度来确定狭长三角形,从而提取到河流等数据空白区域的边缘点。实验结果表明,该算法能够较好地提取LiDAR数据的边缘点,得到LiDAR数据的边缘信息。The definition of triangle shape variable is provided and a novel algorithm based on triangulated irregular network （TIN） for edge detection from airborne LiDAR data is proposed. The TIN is generated in LiDAR data. Each triangle shape variable of TIN is calculated, and the edge triangle is determined according to the different triangle shape variables. These edge triangles are processed to obtain edge points. In view of the LiDAR with blank data area which may be caused by rivers etc., in which case the edge points can not be detected only by the triangle shape variable, using the sum of the quadratic distance from the three vertices to the center of gravity as a measure to determine the skinny triangle is put forward. The experimental results show that the proposed algorithm performs well to extract the edge points and obtain the edge information from the LiDAR data.

关 键 词：遥感 LIDAR数据 三角形形变量 不规则三角网 边缘三角形 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]