• Laser & Optoelectronics Progress
  • Vol. 57, Issue 17, 172801 (2020)
Lei Zhu**, Xingsheng Deng*, Chengbin Xing, and Kang Xu
Author Affiliations
  • Key Laboratory of Highway Engineering of Ministry of Education, Changsha University of Science and Technology, Changsha, Hunan 410114, China
  • show less
    DOI: 10.3788/LOP57.172801 Cite this Article Set citation alerts
    Lei Zhu, Xingsheng Deng, Chengbin Xing, Kang Xu. Moving Surface Filtering Algorithm Based on Multilevel Seed Point Optimization[J]. Laser & Optoelectronics Progress, 2020, 57(17): 172801 Copy Citation Text show less
    Flow chart of the algorithm
    Fig. 1. Flow chart of the algorithm
    Grid index diagram. (a) First-level grid; (b) second-level grid
    Fig. 2. Grid index diagram. (a) First-level grid; (b) second-level grid
    Correspondence between the nodes of the KD tree and the plan sub region
    Fig. 3. Correspondence between the nodes of the KD tree and the plan sub region
    Screening and judgement diagram of seed point. (a) Seed point distribution; (b) slope screening for second-level seed point
    Fig. 4. Screening and judgement diagram of seed point. (a) Seed point distribution; (b) slope screening for second-level seed point
    Points normals and feature plane
    Fig. 5. Points normals and feature plane
    View of points distribution after filtering
    Fig. 6. View of points distribution after filtering
    Comparison of the three types of errors for the two algorithms. (a) Error Ⅰ; (b) error Ⅱ; (c) total error
    Fig. 7. Comparison of the three types of errors for the two algorithms. (a) Error Ⅰ; (b) error Ⅱ; (c) total error
    Comparison with total errors of other eight filtering algorithms
    Fig. 8. Comparison with total errors of other eight filtering algorithms
    DataSampleTerrain featureNumber ofsample pointFirst-levelgrid step /mSlopethreshold /(°)
    Site 1Sample 11Vegetation and buildings on steep slope380103535
    Sample 12Small objects, building on flat ground522193030
    Sample 21Narrow bridge129602530
    Site 2Sample 22Bridges and passages327062530
    Sample 23Complex buildings, discontinuous terrain250954540
    Sample 24Vegetation on the slope74294030
    Site 3Sample 31Low noise point288623030
    Site 4Sample 41Discontinuous terrain112312540
    Sample 42Elongated buildings withhigh-frequency fluctuation terrain424703045
    Site 5Sample 51Vegetation on the slope178453030
    Sample 52Interrupted steep slope224742535
    Sample 53Discontinuous terrain343782535
    Sample 54Village86083030
    Site 6Sample 61Discontinuous terrain, ditch350602540
    Site 7Sample 71Discontinuous terrain, bridge156452535
    Table 1. Data characteristics and initial parameters
    SampleTypeof errorClassicalalgorithmOuralgorithmSampleTypeof errorClassicalalgorithmOuralgorithm
    31.6915.637.778.47
    Sample 1113.5712.65Sample 423.522.88
    Total23.9614.36Total4.784.52
    21.6514.901.721.81
    Sample 1210.559.05Sample 5123.6714.30
    Total15.8212.02Total6.514.54
    1.312.429.136.39
    Sample 2120.039.11Sample 5225.0217.87
    Total5.463.90Total10.807.60
    12.638.0621.1319.35
    Sample 2221.2018.14Sample 5332.8328.01
    Total15.3011.21Total21.6019.70
    30.9817.045.274.16
    Sample 237.907.17Sample 5413.305.92
    Total20.0612.37Total9.585.11
    14.607.935.904.72
    Sample 2424.189.98Sample 6117.587.71
    Total17.318.51Total6.304.82
    22.149.646.444.61
    Sample 3116.7716.20Sample 7127.2925.54
    Total19.6710.92Total8.806.99
    62.3520.0616.989.68
    Sample 416.969.65Average17.6212.95
    Total34.5910.39Total14.709.13
    Table 2. Comparison of three types of errors with classical moving surface algorithm
    Lei Zhu, Xingsheng Deng, Chengbin Xing, Kang Xu. Moving Surface Filtering Algorithm Based on Multilevel Seed Point Optimization[J]. Laser & Optoelectronics Progress, 2020, 57(17): 172801
    Download Citation