基于双目深度筛选的ORB-SLAM3算法

doi:10.13474/j.cnki.11-2246.2024.0113

测绘通报 ›› 2024, Vol. 0 ›› Issue (1): 77-82.doi: 10.13474/j.cnki.11-2246.2024.0113

基于双目深度筛选的ORB-SLAM3算法

符强^1,2,3, 腾先云^1,2,3, 纪元法^1,2,3, 任风华⁴, 孔健明^1,2,3

1. 桂林电子科技大学广西精密导航技术与应用重点实验室, 广西桂林 541004;
2. 桂林电子科技大学信息与通信学院, 广西桂林 541004;
3. 卫星导航与位置服务国家与地方联合工程研究中心, 广西桂林 541004;
4. 桂林电子科技大学电子工程与自动化学院, 广西桂林 541004

收稿日期:2023-04-26 出版日期:2024-01-25 发布日期:2024-01-30
通讯作者: 纪元法。E-mail:1846403892@qq.com
作者简介:符强(1976—),男,硕士,正高级实验师,主要研究方向为图像处理、卫星导航与定位。E-mail:2325950807@qq.com
基金资助:
广西科技厅项目(桂科AA20302022;桂科AB21196041;桂科AB22035074;桂科AD22080061);国家自然科学基金(62061010;62161007);广西八桂学者团队项目;广西高校中青年教师科研基础能力提升项目;(2022KY0181);桂林电子科技大学研究生教育创新计划(2021YCXS026)

ORB-SLAM3 algorithm based on binocular depth screening

FU Qiang^1,2,3, TENG Xianyun^1,2,3, JI Yuanfa^1,2,3, REN Fenghua⁴, KONG Jianming^1,2,3

1. Guangxi Key Laboratory of Precision Navigation Technology and Application, Guilin University of Electronic Technology, Guilin 541004, China;
2. Information and Communicaiton Schnool, Guilin University of Electronic Technology, Guilin 541004, China;
3. National & Local Joint Engineering Research Center of Satellite Navigation Positioning and Location Service, Guilin 541004, China;
4. School of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541004, China

Received:2023-04-26 Online:2024-01-25 Published:2024-01-30

摘要/Abstract

摘要： 针对ORB-SLAM3算法中特征点存在易丢失、精度低,进而导致双目在复杂场景下运动轨迹误差大的问题,本文设计了一种改进的ORB-SLAM3算法。首先,在ORB特征匹配算法中引入自适应角点检测技术,增加特征点的采集数量,并采用光流法跟踪图像特征,提高关键帧的创建成功率;其次,以特征点为中心,作区域搜索,提高实时性;然后,采用双向左右一致性检验筛选最优视差,应用Prosac算法去除误匹配点对;最后,结合深度信息对关键帧进行筛选,提高关键帧的质量,优化相机位姿。采用KITTI和EuRoc数据集进行了试验,验证了改进算法在绝对轨迹误差上具有良好的优化效果。

关键词: 双目视觉, ORB-SLAM3, 光流法, Prosac算法

Abstract: Aiming at the problem that feature points in ORB-SLAM3 algorithm are easily lost and have low accuracy, which in turn leads to large errors in motion trajectory of binoculars in complex scenes, this paper designs an improved ORB-SLAM3 algorithm. Firstly, the adaptive corner point detection technology is introduced in the ORB feature matching algorithm to increase the number of feature point acquisition. Secondly, the optical flow method is used to track the image features to improve the success rate of key frame creation. Then the region search is done with the feature points as the center to improve the real-time performance, the bi-directional left-right consistency test is used to screen the optimal parallax, the Prosac algorithm is applied to remove the mis-matched point pairs. Finally, the depth information is combined with the key frame. the depth information is combined with the key frame screening to improve the quality of key frames and optimize the camera pose. The improved algorithm has good robustness and positioning accuracy in absolute trajectory error.

Key words: binocular vision, ORB-SLAM3, optical flow method, Prosac algorithm

中图分类号:

P237

符强, 腾先云, 纪元法, 任风华, 孔健明. 基于双目深度筛选的ORB-SLAM3算法[J]. 测绘通报, 2024, 0(1): 77-82.

FU Qiang, TENG Xianyun, JI Yuanfa, REN Fenghua, KONG Jianming. ORB-SLAM3 algorithm based on binocular depth screening[J]. Bulletin of Surveying and Mapping, 2024, 0(1): 77-82.

参考文献

[1] 吕德利,王旋.一种基于GPS定位技术的无人机智能光伏巡检系统[J]. 科技创新与应用,2023,13(6): 37-40.
[2] 郭永强,于喜志,黄晓芬,等. UWB定位应用综述[J]. 数字技术与应用,2023,41(2): 32-37.
[3] 胡文龙,周宇飞,宋全军,等. 基于UWB和IMU信息融合的室内定位算法研究[J]. 制造业自动化,2023,45(2): 193-197.
[4] 姚艺,黄卫华,章政,等. 基于改进SIFT的无人机双目目标识别与定位[J]. 组合机床与自动化加工技术,2022(6): 49-53.
[5] DONG J,LIU W,ZHU Y. Application of SIFT feature vector matching algorithm in binocular stereo vision system[C]//Proceedings of the fifth Symposium on Novel Optoelectronic Detection Technology and Application. [S.l.]:SPIE,2019: 437-442.
[6] SHENG H,WEI S,YU X,et al. Research on binocular visual system of robotic arm based on improved SURF algorithm[J]. IEEE Sensors Journal,2019,20(20): 11849-11855.
[7] 杨敬辉,刘德康,杜万和,等.基于图像特征的双目测距系统的研究[J].系统仿真学报,2022,34(3):624-632.
[8] HUO J,ZHOU C,YUAN B,et al. Real-time dense reconstruction with binocular endoscopy based on StereoNet and ORB-SLAM[J]. Sensors,2023,23(4): 2074.
[9] LI C,JIA Y,WANG H,et al. Improved ORB algorithm based on binocular vision[C]//Proceedings of 2019 IEEE International Conference on Computer and Communications (ICCC).[S.l.]:IEEE,2019: 1739-1743.
[10] GOMEZ-OJEDA R,ZUÑIGA-NOL D,MORENO F A,et al. PL-SLAM: a stereo SLAM system through the combination of points and line segments[EB/OL]. 2018-04-09[2023-04-02]. https://arxiv.org/abs/1705.09479.
[11] YAN Y P,WONG S F. A navigation algorithm of the mobile robot in the indoor and dynamicenvironment based on the PF-SLAM algorithm[J]. Cluster Computing,2019,22(6): 14207-14218.
[12] CAMPOS C,ELVIRA R,RODRÍGUEZ J J G,et al. ORB-SLAM3: an accurate open-source library for visual,visual-inertial and multi-map SLAM[J]. IEEE Transactions on Robotics,2021,37(6): 1874-1890.
[13] TEED Z,DENG Jia. DROID-SLAM: deep visual SLAM for monocular,stereo,and RGB-D cameras[EB/OL].2021-08-24[2023-03-30].https://arxiv.org/abs/2108.10869.
[14] 唐浩,张卫强,查云飞,等. 融合超宽带与激光雷达的室内SLAM方法实现[J]. 数据通信,2023(1): 17-23.
[15] 仉新,郑飂默,谭振华,等. 动态场景下移动机器人视觉SLAM[J]. 机床与液压,2023,51(3): 57-63.

基于双目深度筛选的ORB-SLAM3算法

ORB-SLAM3 algorithm based on binocular depth screening

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	黄高爽, 黄桂平, 周杨. 利用光学跟踪与结构光扫描相结合的高精度测量方法[J]. 测绘通报, 2023, 0(4): 135-139.
[2]	岑文广, 许雄, 王国健, 肖长江. 深空探测巡视器视觉导航定位方法研究与地面试验场验证[J]. 测绘通报, 2022, 0(10): 49-55.
[3]	付振华, 邓瑞芝. 一种基于单目视觉的无人机仿地飞行方法[J]. 测绘通报, 2021, 0(11): 128-130.
[4]	卞玉霞, 刘学军, 王美珍, 周业, 甄艳. 双目视觉三维点云重建的误差度量模型[J]. 测绘通报, 2017, 0(11): 64-67.