测绘通报 ›› 2020, Vol. 0 ›› Issue (2): 160-162.doi: 10.13474/j.cnki.11-2246.2020.0065

• 测绘地理信息技术应用案例 • 上一篇    下一篇

改进的自适应遗传算法支持下点云与BIM模型配准

李国远1, 王健1, 刘秀涵2, 孙文潇1   

  1. 1. 山东科技大学测绘学院, 山东 青岛 266590;
    2. 北京麦格天渱科技发展有限公司, 北京 100089
  • 收稿日期:2019-12-02 出版日期:2020-02-25 发布日期:2020-03-04
  • 通讯作者: 王健。E-mail:wangj@sdust.edu.cn E-mail:wangj@sdust.edu.cn
  • 作者简介:李国远(1991-),男,硕士,主要研究方向为三维激光扫描仪及应用。E-mail:395651423@qq.com
  • 基金资助:
    山东省科学基金(ZR2019PD016);山东科技大学研究生科技创新项目(SDKDYC190304)

Point cloud and BIM model registration based on improved adaptive genetic algorithm

LI Guoyuan1, WANG Jian1, LIU Xiuhan2, SUN Wenxiao1   

  1. 1. College of Geomatics, Shandong University of Science and Technology, Qingdao 266590, China;
    2. Beijing Mag Tianhong Technology Development Co., Ltd., Beijing 100089, China
  • Received:2019-12-02 Online:2020-02-25 Published:2020-03-04

摘要: 在钢结构数字化检测中,点云与设计模型的配准是进行数字化检测的关键步骤,配准的精确度决定了检测分析的准确度。传统配准方法一般先进行粗配准再进行精确配准,计算量大且收敛速度缓慢。针对精确配准存在的问题,本文提出了基于改进的自适应遗传算法用于点云与设计模型的配准方法,自适应地调整交叉概率与变异概率的执行顺序及概率值的大小,提高了种群的多样性及收敛速度。试验证明,改进后的自适应遗传算法极大地提高了点云与模型配准精度和收敛速度。

关键词: 点云, BIM模型, 配准, 自适应遗传算法

Abstract: Registration of point cloud and design model is a key step in digital detection of steel structures.The traditional registration method is computationally intensive and slow in convergence. Aiming at accurate registration, an improved adaptive genetic algorithm is proposed for the registration method of point cloud and design model, adaptively adjust the execution order of cross probability and mutation probability and the magnitude' of probability value, and improve the diversity of population and convergence speed. Experiments show that the improved adaptive genetic algorithm greatly improves the accuracy and convergence speed of point cloud and model registration.

Key words: point cloud, BIM model, registration, improved adaptive genetic algorithm

中图分类号: