Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Laser & Optoelectronics Progress, Volume. 60, Issue 22, 2228002(2023)

LiDAR Ground-Segmentation Algorithm Based on Slope Threshold and Convolution Filtering Processing

Tao Shangguan1, Rong Xie1、*, Zufang Lei2, and Zheng Liu1
Author Affiliations
  • 1National Key Laboratory of Radar Signal Processing, Xidian University, Xi'an 710071, Shaanxi , China
  • 2Shenzhen Leishen Intelligent System Co., Ltd., Shenzhen 518100, Guangdong , China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (20)
    Equations (17)
    References (16)
    Tools
    Paper Information
    Topics
    Figures & Tables(20)
    Overall flow chart of the proposed algorithm

    Fig. 1. Overall flow chart of the proposed algorithm

    Download full size
    Schematic of point cloud fan-shaped area division

    Fig. 2. Schematic of point cloud fan-shaped area division

    Download full size
    Angle diagram

    Fig. 3. Angle diagram

    Download full size
    RGB image of the coarsely segmented point cloud

    Fig. 4. RGB image of the coarsely segmented point cloud

    Download full size
    RGB image containing "unknown classification points"

    Fig. 5. RGB image containing "unknown classification points"

    Download full size
    Diagram of guardrail obstacle point cloud and RGB image

    Fig. 6. Diagram of guardrail obstacle point cloud and RGB image

    Download full size
    Diagram of multi-region distance threshold

    Fig. 7. Diagram of multi-region distance threshold

    Download full size
    Diagram of the convolution filtering process

    Fig. 8. Diagram of the convolution filtering process

    Download full size
    RGB image of finely segmented point clouds

    Fig. 9. RGB image of finely segmented point clouds

    Download full size
    Simulated four-wheel vehicle platform

    Fig. 10. Simulated four-wheel vehicle platform

    Download full size
    Simulated road scenes and segmentation effect. (a) Flat road scene; (b) slope road scene; (c) flat road segmentation result; (d) slope road segmentation result

    Fig. 11. Simulated road scenes and segmentation effect. (a) Flat road scene; (b) slope road scene; (c) flat road segmentation result; (d) slope road segmentation result

    Download full size
    Segmentation effect of each stage of the proposed algorithm in different scenes. (a) (c) Results of the first stage; (b) (d) results of the second stage

    Fig. 12. Segmentation effect of each stage of the proposed algorithm in different scenes. (a) (c) Results of the first stage; (b) (d) results of the second stage

    Download full size
    Segmentation effects of three algorithms in different scenes. (a) (d) (g) Method in Ref.[6]; (b) (e) (h) method in Ref.[12]; (c) (f) (i) proposed method

    Fig. 13. Segmentation effects of three algorithms in different scenes. (a) (d) (g) Method in Ref.[6]; (b) (e) (h) method in Ref.[12]; (c) (f) (i) proposed method

    Download full size
    Guardrail physical map

    Fig. 14. Guardrail physical map

    Download full size
    Segmentation effects of the method in Ref.[12] and the proposed method when guardrail obstacles exist. (a) (c) Method in Ref.[12]; (b) (d) proposed method

    Fig. 15. Segmentation effects of the method in Ref.[12] and the proposed method when guardrail obstacles exist. (a) (c) Method in Ref.[12]; (b) (d) proposed method

    Download full size
    Consuming time of three ground segmentation algorithms

    Fig. 16. Consuming time of three ground segmentation algorithms

    Download full size

    • Table 1. Segmentation results of the proposed algorithm in different simulation scenes

      View table

      Table 1. Segmentation results of the proposed algorithm in different simulation scenes

      ScenePprecision /%Precall /%

      Average

      consuming time /ms

      Flat road99.198.210.34
      Slope road98.497.811.28

    • Table 2. Segmentation results for each stage of the proposed algorithm in different scenes

      View table

      Table 2. Segmentation results for each stage of the proposed algorithm in different scenes

      SceneStagePprecision /%Precall /%
      Flat roadThe first stage95.694.5
      Second stage97.596.8
      Complex slope roadThe first stage93.292.3
      Second stage96.195.8

    • Table 3. Segmentation results of the three algorithms in different scenes

      View table

      Table 3. Segmentation results of the three algorithms in different scenes

      SceneMethodPprecision /%Precall /%
      Flat roadMethod in Ref.［6］95.894.7
      Method in Ref.［12］96.895.9
      Proposed method97.596.8
      Complex slope roadMethod in Ref.［6］85.782.3
      Method in Ref.［12］92.491.7
      Proposed method96.195.8

    • Table 4. Average consuming time of three ground segmentation algorithms

      View table

      Table 4. Average consuming time of three ground segmentation algorithms

      ParameterMethod in Ref.［6］Method in Ref.［12］Proposed method
      Averageconsuming time/ms26.6312.2716.57
    Tools

    Get Citation

    Copy Citation Text

    Tao Shangguan, Rong Xie, Zufang Lei, Zheng Liu. LiDAR Ground-Segmentation Algorithm Based on Slope Threshold and Convolution Filtering Processing[J]. Laser & Optoelectronics Progress, 2023, 60(22): 2228002

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Set citation alerts for article
    Save article for my favorites
    Paper Information

    Category: Remote Sensing and Sensors

    Received: Jan. 12, 2023

    Accepted: Feb. 16, 2023

    Published Online: Nov. 6, 2023

    The Author Email: Rong Xie (rxie@mail.xidian.edu.cn)

    DOI:10.3788/LOP230491

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology