东北大学学报:自然科学版 ›› 2020, Vol. 41 ›› Issue (7): 933-937.DOI: 10.12068/j.issn.1005-3026.2020.07.004

• 信息与控制 • 上一篇    下一篇

基于BA优化和KL散度的RGB-D SLAM系统

徐岩, 安卫凤   

  1. (天津大学 电气自动化与信息工程学院, 天津300072)
  • 收稿日期:2019-05-24 修回日期:2019-05-24 出版日期:2020-07-15 发布日期:2020-07-15
  • 通讯作者: 徐岩
  • 作者简介:徐岩(1977-),女,辽宁沈阳人,天津大学副教授.
  • 基金资助:
    国家自然科学基金资助项目(61632018); 青海省基础研究项目(2017-ZJ-753).

RGB-D SLAM System Based on BA Optimization and KL Divergence

XU Yan, AN Wei-feng   

  1. School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China.
  • Received:2019-05-24 Revised:2019-05-24 Online:2020-07-15 Published:2020-07-15
  • Contact: AN Wei-feng
  • About author:-
  • Supported by:
    -

摘要: 针对深度相机采集深度图像的噪声对位姿估计精度的影响,以及误差随时间累积的严重问题,设计了一种改进的基于RGB-D相机的视觉SLAM系统.首先,建立重投影误差模型,通过最小化重投影误差,对位姿和特征点进行非线性优化.此外,提出了一种闭环检测的算法,建立字典模型,用频率-逆文档频率计算权重,用Kullback-Leibler散度计算相似度,并使用相对相似度机制检测闭环,减少了累积误差.使用15个公开的图像序列对算法进行评价,同3个流行的RGB-D SLAM系统对比,精度平均最高提高119.07%,最低提高4.24%.实验结果证明,提出的方法比目前流行的RGB-D SLAM系统具有更好的精度.

关键词: 机器视觉, 同时定位与建图, BA优化, KL散度, 闭环检测

Abstract: To reduce the impact of noise in depth image acquisition on the accuracy of pose estimation and to solve the serious problem of cumulative error over time, an improved RGB-D SLAM system was designed. Firstly, the re-projection error model was established to nonlinearly optimize the poses and features by minimizing the re-projection error. In addition, a closed-loop detection algorithm was proposed. The dictionary model was established and the frequency-inverse document frequency(TF-IDF) was used to calculate the weight. Kullback-Leibler divergence was used to calculate the similarity, and a relative similarity mechanism was used for the closed-loop detection. The cumulative error was decreased. The algorithm was evaluated using 15 public image sequences. Compared with three popular RGB-D SLAM systems, the maximum increase of accuracy is averagely 119.07% and the minimum increase is 4.24%. Experimental results show that the proposed method has better accuracy than the current popular RGB-D SLAM systems.

Key words: machine vision, simultaneous localization and mapping, BA optimization, KL divergence, closed-loop detection

中图分类号: