• Laser & Optoelectronics Progress
  • Vol. 59, Issue 12, 1228008 (2022)
Changyong Zhang and Liang Han*
Author Affiliations
  • College of Electronic Information and Automation, Civil Aviation University of China, Tianjin 300300, China
  • show less
    DOI: 10.3788/LOP202259.1228008 Cite this Article Set citation alerts
    Changyong Zhang, Liang Han. Obstacle Detection of LiDAR Based on Optimized DBSCAN[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1228008 Copy Citation Text show less
    Schematic of ground points in different scenes. (a) Schematic of slope ground point determination; (b) schematic of non-ground point judgment
    Fig. 1. Schematic of ground points in different scenes. (a) Schematic of slope ground point determination; (b) schematic of non-ground point judgment
    Schematic of 3D LiDAR scanning. (a) Side view of 3D LiDAR scanning; (b) front view of 3D LiDAR scanning
    Fig. 2. Schematic of 3D LiDAR scanning. (a) Side view of 3D LiDAR scanning; (b) front view of 3D LiDAR scanning
    Flowchart of obstacle detection
    Fig. 3. Flowchart of obstacle detection
    Selection of representative points
    Fig. 4. Selection of representative points
    ROI data extraction and ground segmentation results. (a) Lane line detection; (b) ROI extraction; (c) ground segmentation
    Fig. 5. ROI data extraction and ground segmentation results. (a) Lane line detection; (b) ROI extraction; (c) ground segmentation
    Multi-distance and multi-obstacle plane scene detection experiment. (a) Data acquisition scenarios; (b) effect of traditional DBSCAN algorithm; (c) effect of algorithm in Ref. [9]; (d) effect of proposed algorithm
    Fig. 6. Multi-distance and multi-obstacle plane scene detection experiment. (a) Data acquisition scenarios; (b) effect of traditional DBSCAN algorithm; (c) effect of algorithm in Ref. [9]; (d) effect of proposed algorithm
    Multi-distance and multi-obstacle slope scene detection experiment. (a) Data acquisition scenarios; (b) effect of traditional DBSCAN algorithm; (c) effect of algorithm in Ref. [9]; (d) effect of proposed algorithm
    Fig. 7. Multi-distance and multi-obstacle slope scene detection experiment. (a) Data acquisition scenarios; (b) effect of traditional DBSCAN algorithm; (c) effect of algorithm in Ref. [9]; (d) effect of proposed algorithm
    MaterialReflectance /%
    Asphalt7
    Cement31
    Concrete40
    Fresh yellow paint55
    Fresh white paint60
    Table 1. Reflection intensity of different materials
    SceneAlgorithm

    Positive

    recognition

    Error

    recognition

    MissedrecognitionPositiverecognition rate /%Averagetime /s
    Straight roadTraditional DBSCAN algorithm221258766.371.26
    Ref.[9275481282.090.12
    Improved DBSCAN algorithm299241089.790.08
    Sloping roadTraditional DBSCAN algorithm195318462.901.29
    Ref.[9246832170.290.17
    Improved DBSCAN algorithm297291387.610.11
    Table 2. Performance comparison of traditional DBSCAN algorithm, Ref. [9] algorithm, and proposed algorithm
    Changyong Zhang, Liang Han. Obstacle Detection of LiDAR Based on Optimized DBSCAN[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1228008
    Download Citation