基于Deeplabv3+和LK光流的动态视觉SLAM算法
DOI:
CSTR:
作者:
作者单位:

1.北京信息科技大学机电工程学院 北京 100192;2.新能源汽车北京实验室 北京 100192

作者简介:

通讯作者:

中图分类号:

TP391.9;TN98

基金项目:

国家自然科学基金面上项目(52077007)资助


Dynamic visual SLAM algorithm based on deeplabv3+ and LK optical flow
Author:
Affiliation:

1.School of Mechanical and Electrical Engineering, Beijing Information Science and Technology University, Beijing 100192, China;2.New Energy Vehicle Research Lab,Beijing 100192,China

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:

    传统ORB-SLAM3系统在静态环境中表现优秀,但存在动态特征时会引入不必要的噪声,造成特征匹配出现错误,而现有动态SLAM算法难以完整判断潜在动态特征,出现漏检或误检导致定位精度下降。针对这些问题,将语义分割网络Deeplabv3+与LK光流法融合进ORB-SLAM3的跟踪线程,首先将Deeplabv3+的主干网络替换为Mobilenetv3,提高语义分割的精度,然后使用语义分割获取潜在动态目标掩码,初步滤除动态特征点,对剩余特征点进行LK光流计算,将光流平均误差作为阈值防止剩余静态特征点过少导致位姿估计失败。相比于原ORB-SLAM3,本文改进的算法在TUM高动态序列的定位精度平均提升了47.92%,与现有优秀动态SLAM算法相比,本文算法在TUM数据集的Walking_static序列取得了最高的定位精度。

    Abstract:

    The traditional ORB-SLAM3 system demonstrates excellent performance in static environments, however, the presence of dynamic features introduces unnecessary noise, leading to errors in feature matching and inaccuracies in camera pose estimation. Existing dynamic SLAM algorithms face challenges in comprehensively identifying potential dynamic features, resulting in missed detections or false positives and consequently degrading localization accuracy. To tackle these issues, the semantic segmentation network Deeplabv3+ and the Lucas-Kanade optical flow method are incorporated into the tracking thread of ORB-SLAM3. Specifically, the backbone network of Deeplabv3+ is replaced with Mobilenetv3 to enhance the precision of semantic segmentation. Semantic segmentation is then used to obtain a mask of potential dynamic objects, which is employed to preliminarily filter out dynamic feature points. The remaining feature points undergo LK optical flow calculation, with the average optical flow error serving as a threshold to prevent the insufficient number of static feature points from causing pose estimation failure. In comparison to the original ORB-SLAM3, the improved algorithm in this study achieves an average localization accuracy improvement of 47.92% on the high-dynamic sequences of the TUM dataset. Furthermore, among existing advanced dynamic SLAM algorithms, the proposed method achieved the highest localization accuracy on the Walking_static sequence of the TUM dataset.

    参考文献
    相似文献
    引证文献
引用本文

周若轩,张瑞乾,陈勇,袁旭浩,秦慧军.基于Deeplabv3+和LK光流的动态视觉SLAM算法[J].电子测量技术,2025,48(9):149-155

复制
相关视频

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:
  • 录用日期:
  • 在线发布日期: 2025-05-23
  • 出版日期:
文章二维码