测绘通报 ›› 2019, Vol. 0 ›› Issue (11): 89-92.doi: 10.13474/j.cnki.11-2246.2019.0358

• 学术研究 • 上一篇    下一篇

DBSCAN聚类和改进的双边滤波算法在点云去噪中的应用

曲金博, 王岩, 赵琪   

  1. 沈阳建筑大学交通工程学院, 辽宁 沈阳 110168
  • 收稿日期:2018-12-29 发布日期:2019-12-02
  • 作者简介:曲金博(1994-),男,硕士生,研究方向为精密工程测量和三维激光扫描技术。E-mail:3281927237@qq.com
  • 基金资助:
    国家自然科学基金(51774204)

Application of DBSCAN clustering and improved bilateral filtering algorithm in point cloud denoising

QU Jinbo, WANG Yan, ZHAO Qi   

  1. School of Transportation Engineering, Shenyang Jianzhu University, Shenyang 110168, China
  • Received:2018-12-29 Published:2019-12-02

摘要: 采用基于密度的DBSCAN聚类算法对点云数据进行去噪处理,然后通过改进的双边滤波方法进行光顺处理实现点云平滑效果,最终的结果不仅有效去除了噪声点,还保留了点云模型的特征。以沈阳民国时期代表性的建筑——沈阳金融博物馆为试验模型进行试验,结果表明:通过DBSCAN聚类算法处理后得到的点云数据,再经改进的双边滤波处理所得到的数据远远比原点云数据直接运用改进的双边滤波处理得到的数据精度高,点云去噪效果更好。

关键词: DBSCAN聚类算法, 双边滤波方法, 噪声点, 点云, 密度

Abstract: The density-based DBSCAN clustering algorithm is used to denoise the point cloud data, and the smoothing effect is achieved by the improved bilateral filtering method that conducts smooth treatment. Finally not only the noise points are effectively removed, but also characteristics of the point cloud model are retained. This article uses the representative building of Shenyang during the Republic of China-Shenyang Financial Museum as the experimental model. The experimental results show that the point cloud data obtained by the DBSCAN clustering algorithm and the improved bilateral filtering process are far more accurate than the original point cloud data, and the data is more accurate and denoising, point cloud denoising is better.

Key words: DBSCAN clustering algorithm, Bilateral filtering method, Noise point, point cloud, density

中图分类号: