• Optics and Precision Engineering
  • Vol. 32, Issue 19, 2945 (2024)
Kuiyu ZHOU1, Yuchun HUANG1,*, He YANG2, and Na LI3
Author Affiliations
  • 1School of Remote Sensing and Information Engineering, Wuhan University, Wuhan430079, China
  • 2Transportation Development Center of Henan Province, Zhengzhou450000, China
  • 3Troops 61175 of PLA, Nanjing210049, China
  • show less
    DOI: 10.37188/OPE.20243219.2945 Cite this Article
    Kuiyu ZHOU, Yuchun HUANG, He YANG, Na LI. Integrated 2D-3D LiDAR-vision fusion vehicle speed estimation based on image frustum[J]. Optics and Precision Engineering, 2024, 32(19): 2945 Copy Citation Text show less
    Flowchart of fusion-based vehicle speed estimation technique
    Fig. 1. Flowchart of fusion-based vehicle speed estimation technique
    Differentiating foreground from background using ground points
    Fig. 2. Differentiating foreground from background using ground points
    Preliminary segmentation of main point cloud and noise point cloud in overhead 2D view and selection of seed point regions
    Fig. 3. Preliminary segmentation of main point cloud and noise point cloud in overhead 2D view and selection of seed point regions
    Restoration to 3D point cloud and selection of clustering seed points based on centroids
    Fig. 4. Restoration to 3D point cloud and selection of clustering seed points based on centroids
    Traffic conditions in various scenarios
    Fig. 5. Traffic conditions in various scenarios
    Visualization of calibration results in some scenarios
    Fig. 6. Visualization of calibration results in some scenarios
    Speed measurement results of descending after straight driving from near to far
    Fig. 7. Speed measurement results of descending after straight driving from near to far
    Speed measurement results of downhill and straight driving from far to near
    Fig. 8. Speed measurement results of downhill and straight driving from far to near
    Results of vehicles traveling from far to near and turning at a gentle corner at the end
    Fig. 9. Results of vehicles traveling from far to near and turning at a gentle corner at the end
    Speed measurement results of vehicles approaching a stop state
    Fig. 10. Speed measurement results of vehicles approaching a stop state
    Speed measurement results of vehicles during turning
    Fig. 11. Speed measurement results of vehicles during turning
    Image-prioritized multi-target vehicle extraction and localization results
    Fig. 12. Image-prioritized multi-target vehicle extraction and localization results
    场景重投影误差
    10.023 5
    20.008 9
    30.011 6
    40.021 3
    Table 1. Reprojection error after calibration of each scenario
    行驶状态正常车流量密集车流量

    平均绝对误差

    MAE/(m·s-1

    均方根误差

    RMSE/(m·s-1

    平均绝对误差

    MAE/(m·s-1

    均方根误差

    RMSE/(m·s-1

    直行0.2070.2560.2350.277
    启停0.2670.2750.2840.330
    转弯0.3120.3520.3090.368
    Table 2. Average speed measurement error of car
    行驶状态正常车流量
    平均绝对误差MAE/(m·s-1

    均方根误差RMSE

    /(m·s-1

    直行0.1830.246
    启停0.2260.251
    转弯0.2810.316
    Table 3. Average speed measurement error of passenger car, truck and lorry
    Kuiyu ZHOU, Yuchun HUANG, He YANG, Na LI. Integrated 2D-3D LiDAR-vision fusion vehicle speed estimation based on image frustum[J]. Optics and Precision Engineering, 2024, 32(19): 2945
    Download Citation