###
DOI:
电力大数据:2024,27(4):-
←前一篇   |   后一篇→
本文二维码信息
基于改进ORBSLAM的三维点云重建方法
(1.贵州电网有限公司电力科学研究院;2.贵州大学电力科学研究院工作站)
A 3D point cloud reconstruction method based on improved ORBSLAM
LIU Xing1,2, BAN Guobang1,2, WU Hao1,2, MENG Lingwen3,4,5,2, ZHOU Junchao1,2
(1.Electric Power Research Institute of?2.Guizhou Power Grid Co,Ltd;3.Electric Power Research Institute of&4.amp;5.#160)
摘要
图/表
参考文献
相似文献
本文已被:浏览 48次   下载 46
投稿时间:2024-06-24    修订日期:2024-07-09
中文摘要: 针对电网复杂环境特点,研发适用的三维成像装置,实现对作业场景的全方位、高精度的成像。该成果能够提供更全面的环境感知和风险监测,为立体防控提供可靠的基础数据。为此,本文提出了一种基于改进ORBSLAM的三维点云重建方法。首先,选用深度相机作为实验设备来获取点云数据的像素坐标,然后,通过像素坐标到图像坐标的变换,转换成三维点云数据,最后,使用简单的松耦合改进 ORBSLAM算法,补充估计的相机位姿缺失以及检测估计偏差较大的误差,提高作业场景三维重建点云融合准确率。
Abstract:Based on the complex environmental characteristics of the power grid, a suitable 3D imaging device is developed to achieve all-round and high-precision imaging of the work scene. This achievement can provide more comprehensive environmental perception and risk monitoring, and provide reliable basic data for three-dimensional prevention and control. Therefore, this article proposes a 3D point cloud reconstruction method based on improved ORBSLAM. Firstly, a depth camera is selected as the experimental equipment to obtain the pixel coordinates of point cloud data. Then, through the transformation from pixel coordinates to image coordinates, it is converted into three-dimensional point cloud data. Finally, a simple loosely coupled improved ORBSLAM algorithm is used to supplement the estimated camera pose and detect errors with large estimation deviations, improving the accuracy of point cloud fusion in three-dimensional reconstruction of work scenes.
文章编号:     中图分类号:    文献标志码:
基金项目:复杂作业人身风险态势感知与立体防控技术研究(GZKJXM20222338)
引用文本: