| 激光雷达的多源数据融合点云分类算法研究 |
投稿时间：2022-06-29 修订日期：2022-09-07  点此下载全文 |
| 引用本文： |
| 摘要点击次数: 64 |
| 全文下载次数: 0 |
|
| 基金项目:基金项目：江西教育厅科学技术研究一般项目(GJJ211031) |
|
| 中文摘要:摘要：随机采样一致性算法（Random Consistent Sampling，RANSAC）是常见的激光点云分割方法，但该算法推广至输电线场景时存在点云分割效率低、抗噪性差等不足，不利于高精度的输电线模型重建及后续线路风险检测。根据单档输电线空间分布特性，提出了改进随机采样一致的输电线点云分割方法。首先优化初始样本点选择原则、引入最小二乘原理参数求解等改进策略，提高了随机采样一致性算法输电线模型重建精度；然后以直线-抛物线方程为单根输电线识别的约束条件，利用逐根提取方式实现输电线激光点云分割。选择两组典型代表性的机载激光点云数据实验分析，结果表明：该方法有效解决了数据缺失、点云噪声等复杂背景环境的输电线激光点云分割，准确率、召回率和整体精度最小值分别为99.19％，99.25％，99.10％。较之已有方法，具有点云分割精度高、算法普适性强的优势。 |
| 中文关键词:关键词：机载激光雷达，改进随机一致性采样，输电线，点云分割 |
| |
| Research on point cloud classification algorithm of multi-source data fusion for lidar |
|
|
| Abstract:Abstract: Random Consistent Sampling (RANSAC) is a common method for laser point cloud segmentation, but it has some shortcomings such as low segmentation efficiency and poor anti-noise performance when it is extended to transmission line scenes, transmission Line Model Reconstruction and subsequent line risk detection which is unfavorable to high accuracy. According to the spatial distribution characteristics of single-span transmission lines, an improved point cloud segmentation method based on uniform random sampling is proposed. Firstly, the principle of initial sample selection is optimized and the parameters of least square method is introduced to improve the reconstruction accuracy of the transmission line model Then the line-parabola equation is taken as the constraint condition of single transmission line identification, and the laser point cloud segmentation of transmission line is realized by extracting line by line. Two groups of typical airborne laser point cloud data were selected and analyzed. The results show that this method can effectively solve the problem of laser point cloud segmentation in complex background environment such as data missing and point cloud noise, the minimum values of accuracy, recall and overall accuracy were 99.19% , 99.25% and 99.10% , respectively. Compared with the existing methods, the algorithm has the advantages of high precision of point cloud segmentation and strong universality. |
| keywords:关键词：机载激光雷达,改进随机一致性采样,输电线,点云分割 |
| 查看全文 查看/发表评论 下载pdf阅读器 |
次