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

Chinese Journal of Lasers, Volume. 47, Issue 9, 910002(2020)

Detection and Geo-localization of Small Traffic Signs Based on Images and Laser Data

Liu Lirong1,2、*, Tang Xinming2, Zhao Wenji1, Gao Xiaoming2, and Xie Junfeng2
Author Affiliations
  • 1College of Resource Environment and Tourism, Capital Normal University, Beijing 100048, China
  • 2Land Satellite Remote Sensing Application Center, MNR, Beijing 100048, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (12)
    Equations (0)
    References (22)
    Cited By (3)
    Tools
    Paper Information
    Topics
    Figures & Tables(12)
    Schematic of overall plan

    Fig. 1. Schematic of overall plan

    Download full size
    Object detection process based on SlimNet model

    Fig. 2. Object detection process based on SlimNet model

    Download full size
    Three types of coordinate systems

    Fig. 3. Three types of coordinate systems

    Download full size
    3D space coordinates of intersection

    Fig. 4. 3D space coordinates of intersection

    Download full size
    Process of spatial geolocation based on depth map

    Fig. 5. Process of spatial geolocation based on depth map

    Download full size
    Process of extracting diagonal of target. (a) Detect target; (b) determine endpoint; (c) extract result

    Fig. 6. Process of extracting diagonal of target. (a) Detect target; (b) determine endpoint; (c) extract result

    Download full size
    Annotation examples of image targets

    Fig. 7. Annotation examples of image targets

    Download full size
    Results of initial vectorization

    Fig. 8. Results of initial vectorization

    Download full size
    Examples of traffic sign geolocation and vectorization results

    Fig. 9. Examples of traffic sign geolocation and vectorization results

    Download full size

    • Table 1. Statistical results of experimental data

      View table

      Table 1. Statistical results of experimental data

      ClassTrainingTestBackground
      LCZ816272
      Circle47715815000
      Triangle11338

    • Table 2. Comparison of detection results between SlimNet model and VGG16 model

      View table

      Table 2. Comparison of detection results between SlimNet model and VGG16 model

      ModelNumber ofiterationsLearningrateTest1Test2 (Test1+Background)
      APmAP /%APmAP /%
      LCZCircleTriangleLCZCircleTriangle
      SlimNet3.5×1041.5×1040.00050.0001100.096.696.597.899.862.863.575.4
      VGG163.5×1041.5×1040.00010.000199.395.196.396.996.960.456.371.2

    • Table 3. Three types of target geolocation and vectorization results statistics

      View table

      Table 3. Three types of target geolocation and vectorization results statistics

      Test areaClassFPTPFNRecall /%Precision /%
      Shi ZuishanLCZ251198.196.2
      Shen ZhenCircle759986.789.4
      Triangle531488.686.1
    Tools

    Get Citation

    Copy Citation Text

    Liu Lirong, Tang Xinming, Zhao Wenji, Gao Xiaoming, Xie Junfeng. Detection and Geo-localization of Small Traffic Signs Based on Images and Laser Data[J]. Chinese Journal of Lasers, 2020, 47(9): 910002

    Download Citation

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

    Category: remote sensing and sensor

    Received: Mar. 25, 2020

    Accepted: --

    Published Online: Sep. 16, 2020

    The Author Email: Lirong Liu (liulirong1125@163.com)

    DOI:10.3788/CJL202047.0910002

    Topics

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