PFD-SLAM: A New RGB-D SLAM for Dynamic Indoor Environments Based on Non-Prior Semantic Segmentation

被引:21
|
作者
Zhang, Chenyang [1 ]
Zhang, Rongchun [2 ,3 ]
Jin, Sheng [4 ]
Yi, Xuefeng [5 ]
机构
[1] Changzhou Inst Technol, Sch Civil Engn & Architecture, Changzhou 213032, Jiangsu, Peoples R China
[2] Nanjing Univ Posts & Telecommun, Sch Geog & Biol Informat, Nanjing 210023, Peoples R China
[3] Nanjing Univ Posts & Telecommun, Sch Commun & Informat Engn, Nanjing 210023, Peoples R China
[4] Tianjin Univ, Inst Robot & Autonomous Syst, Sch Elect & Informat Engn, Tianjin 300072, Peoples R China
[5] Hohai Univ, Sch Earth Sci & Engn, Nanjing 211000, Peoples R China
基金
中国博士后科学基金; 中国国家自然科学基金;
关键词
RGB-D SLAM; feature-level segmentation; dynamic scenes; non-prior semantic segmentation; particle filter; VISUAL ODOMETRY; LOCALIZATION; ROBUST;
D O I
10.3390/rs14102445
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Now, most existing dynamic RGB-D SLAM methods are based on deep learning or mathematical models. Abundant training sample data is necessary for deep learning, and the selection diversity of semantic samples and camera motion modes are closely related to the robust detection of moving targets. Furthermore, the mathematical models are implemented at the feature-level of segmentation, which is likely to cause sub or over-segmentation of dynamic features. To address this problem, different from most feature-level dynamic segmentation based on mathematical models, a non-prior semantic dynamic segmentation based on a particle filter is proposed in this paper, which aims to attain the motion object segmentation. Firstly, GMS and optical flow are used to calculate an inter-frame difference image, which is considered an observation measurement of posterior estimation. Then, a motion equation of a particle filter is established using Gaussian distribution. Finally, our proposed segmentation method is integrated into the front end of visual SLAM and establishes a new dynamic SLAM, PFD-SLAM. Extensive experiments on the public TUM datasets and real dynamic scenes are conducted to verify location accuracy and practical performances of PFD-SLAM. Furthermore, we also compare experimental results with several state-of-the-art dynamic SLAM methods in terms of two evaluation indexes, RPE and ATE. Still, we provide visual comparisons between the camera estimation trajectories and ground truth. The comprehensive verification and testing experiments demonstrate that our PFD-SLAM can achieve better dynamic segmentation results and robust performances.
引用
收藏
页数:24
相关论文
共 50 条
  • [1] Semantic Segmentation based Dense RGB-D SLAM in Dynamic Environments
    Zhang, Jianbo
    Liu, Yanjie
    Chen, Junguo
    Ma, Liulong
    Jin, Dong
    Chen, Jiao
    2019 3RD INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, AUTOMATION AND CONTROL TECHNOLOGIES (AIACT 2019), 2019, 1267
  • [2] Improving RGB-D SLAM in dynamic environments using semantic aided segmentation
    Kenye, Lhilo
    Kala, Rahul
    ROBOTICA, 2022, 40 (06) : 2065 - 2090
  • [3] RS-SLAM: A Robust Semantic SLAM in Dynamic Environments Based on RGB-D Sensor
    Ran, Teng
    Yuan, Liang
    Zhang, Jianbo
    Tang, Dingxin
    He, Li
    IEEE SENSORS JOURNAL, 2021, 21 (18) : 20657 - 20664
  • [4] A robust RGB-D SLAM based on multiple geometric features and semantic segmentation in dynamic environments
    Kuang, Benfa
    Yuan, Jie
    Liu, Qiang
    MEASUREMENT SCIENCE AND TECHNOLOGY, 2023, 34 (01)
  • [5] RGB-D SLAM in Dynamic Environments with Multilevel Semantic Mapping
    Qin, Yusheng
    Mei, Tiancan
    Gao, Zhi
    Lin, Zhipeng
    Song, Weiwei
    Zhao, Xuhui
    JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS, 2022, 105 (04)
  • [6] RGB-D SLAM in Dynamic Environments with Multilevel Semantic Mapping
    Yusheng Qin
    Tiancan Mei
    Zhi Gao
    Zhipeng Lin
    Weiwei Song
    Xuhui Zhao
    Journal of Intelligent & Robotic Systems, 2022, 105
  • [7] MG-SLAM: RGB-D SLAM Based on Semantic Segmentation for Dynamic Environment in the Internet of Vehicles
    Zhang, Fengju
    Zhu, Kai
    CMC-COMPUTERS MATERIALS & CONTINUA, 2025, 82 (02): : 2353 - 2372
  • [8] An RGB-D SLAM algorithm based on adaptive semantic segmentation in dynamic environment
    Song Wei
    Zhang Li
    Journal of Real-Time Image Processing, 2023, 20
  • [9] An RGB-D SLAM algorithm based on adaptive semantic segmentation in dynamic environment
    Wei, Song
    Li, Zhang
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2023, 20 (05)
  • [10] An RGB-D SLAM Algorithm Based on Adaptive Semantic Segmentation in Dynamic Environment
    Wang M.
    Song W.
    Jiqiren/Robot, 2023, 45 (01): : 16 - 27