Research on Loop Closing for SLAM Based on RGB-D Images

  • Hongwei MoEmail author
  • Kai Wang
  • Haoran Wang
  • Weihao Ding
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 528)


This paper mainly studies a loop closing detection method based on visual SLAM. We used RGB-D image as data source. The main idea is to construct a word bag based on DBoW3. Using rBRIEF makes it possible to perform feature extraction after the image is rotated. And added the elimination of mis-match links to improve the accuracy of detection. In order to ensure the reliability of the loop closing test results, the matching image is also verified. RGB-D image is rich in information and can synchronously extract the depth and color information of the main objects in the scene. The depth information directly reflects the distance information of each object in the scene.


SLAM Loop closing RGB-D DBoW3 


  1. 1.
    J. Sivic, A. Zisserman, Video google: a text retrieval approach to object matching in videos, in 2003 IEEE International Conference on Computer Vision, vol. 2 (Nice, France, 2003), p. 1470Google Scholar
  2. 2.
    P. Newman, K. Ho, SLAM-loop closing with visually salient features, in Proceedings of the 2005 IEEE International Conference on Robotics and Automation (Washington, America, 2005), pp. 635–642Google Scholar
  3. 3.
    D. Nister, H. Stewenius, Scalable recognition with a vocabulary tree, in IEEE Computer Society Conference on Computer Vision and Pattern Recognition (New York, America, 2006), pp. 2161–2168Google Scholar
  4. 4.
    A. Angeli, D. Filliat, Fast and incremental method for loop-closure detection using bags of visual words. IEEE Trans. Rob. 24(5), 1027–1037 (2008)CrossRefGoogle Scholar
  5. 5.
    C. Cadena, D. Galvez-Lopez, Robust place recognition with stereo cameras, in IEEE/RSJ International Conference on Intelligent Robots and Systems, vol. 25, no. 1 (Taipei, Taiwan 2010), pp. 5182–5189Google Scholar
  6. 6.
    Y. Liu, H. Zhang, Visual loop closure detection with a compact image descriptor, in 2012 IEEE/RSJ International Conference on Intelligent Robots & Systems, vol. 57, no. 1 (Vilamoura, Algarve, 2012), pp. 1051–1056Google Scholar
  7. 7.
    M. Cummins, P. Newman, Probabilistic appearance based navigation and loop closing, in IEEE International Conference on Robots & Automation (Roma, Italy, 2007), pp. 2042–2048Google Scholar
  8. 8.
    M. Cummins, P. Newman, Appearance-only SLAM at large scale with FAB-MAP 2.0. Int. J. Robot. Res. 30(9), 1100–1123 (2011)CrossRefGoogle Scholar
  9. 9.
    S. Tully, G. Kantor, A unified bayesian framework for global localization and SLAM in hybrid metric topological maps. Int. J. Robot. Res. 31(31), 271–288 (2012)CrossRefGoogle Scholar
  10. 10.
    F.M. Campos, L. Correia, Mobile robot global localization with non-quantized SIFT features, in IEEE 15th International Conference on Advanced Robotics: New Boundaries for Robotics (Tallin, 2011), pp. 582–587Google Scholar
  11. 11.
    R. Mur-Artal, J.D. Tardos, ORB-SLAM2: an open-source system for monocular, stereo, and RGB-D cameras. IEEE Trans. Rob. PP(99), 1–8 (2016)Google Scholar
  12. 12.
    C. Kerl, J. Sturm, Dense visual SLAM for RGB-D cameras, in 2014 IEEE/RSJ International Conference on Intelligent Robots & Systems, vol. 8215, no. 2 (Chicago, America, 2014), pp. 2100–2106Google Scholar
  13. 13.
    Y. Latif, C. Cadena, Robust loop closing over time for pose graph SLAM. Int. J. Rob. Res. 32(14), 1611–1626 (2013)CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • Hongwei Mo
    • 1
    Email author
  • Kai Wang
    • 1
  • Haoran Wang
    • 1
  • Weihao Ding
    • 1
  1. 1.School of AutomationHarbin Engineering UniversityHarbinChina

Personalised recommendations