基于改进Delaunay算法的树冠三维重构单木因子提取

Amended Delaunay Algorithm for Single Tree Factor Extraction Using 3-D Crown Modeling

摘要:

针对树冠结构复杂和点云数据量大的特点，为了提高测算单木因子中树冠表面积和体积的精度和效率，通过改进Delaunay三角网的算法机制，提出一种基于空间分割的分块优先级机制的三角网表面重建算法，用于重构树冠表面，形成一种高精度高效率的树冠表面积和体积因子提取方法。利用地面三维激光扫描系统获取树冠点云数据，分别通过传统人工方法、点云量测法、数字高程模型算法与提出的改进SD Delaunay算法，计算实验区域研究对象的树冠表面积与体积，并进行对比分析，结果表明提出的算法完全满足计算精度，同时该计算方法的耗时只有传统人工法的41%，数字高程模型法的62%，大大提高了运算效率。

Abstract:

Crown structure is complex and has huge amount of point cloud data. In order to achieve more accurate and more efficient measurement of crown surface area and volume, the characteristics of existing algorithms were analyzed and an improved triangular mesh surface reconstruction algorithm which is based on spatial division block priority mechanism was proposed for the reconfiguration of the crown surface. A precise and efficient factor extraction method for crown surface area and volume was formed. Crown point cloud data obtained through ground 3-D laser scanning system was used. Through the traditional artificial method, cloud point measurement method, the classics digital elevation model algorithm and the proposed algorithm, the crown surface and volume of the experimental subjects was calculated respectively. The inter comparison of the algorithm results fully meets the calculation accuracy. The time consumption of the proposed algorithm only accounts for 41% of the traditional way and 62% of digital elevation model method, which appears to greatly improve the operation efficiency. 