基于激光雷达及特征匹配的室内场景设计重建
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

河北省自然科学基金项目(No.20204738275);河北省科技计划项目(No.2020-06X)资助。


Indoor scene design reconstruction based on lidar and feature matching
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对当前采集到的场景出现缺损导致场景设计重建效率较低、准确性较差的问题,提出基于激光雷达扫描及关键点特征匹配的室内场景设计重建。通过Deleta 2B型激光雷达传感器直接采集室内场景图像,利用最小二乘拟合滤波算法曲线拟合室内场景图像的灰度值,根据图像特征向量匹配室内场景图像的关键点特征,通过德洛内三角化算法串联室内场景点云数据,构造三维网格模型,通过wallis滤波器提升重建辨识度,实现室内场景的重建。实验结果表明,该方法重建后的室内图像可辨识度较高,信噪比最高可达到498dB,室内场景图像覆盖率均在90以上,重建场景的结构相似度接近1,次卧室场景重建时间为592 h。

    Abstract:

    Aiming at the problem of low efficiency and poor accuracy of scene design reconstruction due to defects in the currently collected scenes,an indoor scene design reconstruction based on lidar scanning and key point feature matching is proposed.The indoor scene image is directly collected through the Deleta 2B lidar sensor,and the gray value of the indoor scene image is curve fitted by the least squares fitting filter algorithm,and the key point features of the indoor scene image are matched according to the image feature vector.The triangulation algorithm connects the indoor scene point cloud data to construct a three dimensional grid model.The wallis filter is used to improve the recognition of the reconstruction and realize the reconstruction of the indoor scene.The experimental results show that the reconstructed indoor images of the method are highly identifiable,the signal to noise ratio can reach up to 49.8 dB,the image coverage of indoor scenes are above 90%,the structural similarity of the reconstructed scenes is close to 1,and the reconstruction time of the sub bedroom scenes is 59.2 h.

    参考文献
    相似文献
    引证文献
引用本文

谷晓龙,张文松.基于激光雷达及特征匹配的室内场景设计重建[J].激光与红外,2022,52(9):1335~1341
GU Xiao-long, ZHANG Wen-song. Indoor scene design reconstruction based on lidar and feature matching[J]. LASER & INFRARED,2022,52(9):1335~1341

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:2021-11-15
  • 录用日期:
  • 在线发布日期: 2022-09-23
  • 出版日期: