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

Acta Optica Sinica, Volume. 39, Issue 8, 0815006(2019)

Three-Dimensional Object Recognition Based on Enhanced Point Pair Features

Rongrong Lu1,2,3,4,5、**, Feng Zhu1,2,4,5、*, Qingxiao Wu1,2,4,5, Foji Chen1,2,3,4,5, Yunge Cui1,2,3,4,5, and Yanzi Kong1,2,3,4,5
Author Affiliations
  • 1 Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, Liaoning 110016, China
  • 2 Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang, Liaoning 110016, China
  • 3 University of Chinese Academy of Sciences, Beijing 100049, China
  • 4 Key Laboratory of Opto-Electronic Information Process, Chinese Academy of Sciences, Shenyang, Liaoning 110016, China
  • 5 Key Laboratory of Image Understanding and Computer Vision, Shenyang, Liaoning 110016, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (15)
    Equations (0)
    References (12)
    Cited By (1)
    Tools
    Paper Information
    Topics
    Figures & Tables(15)
    Schematic of ambiguity of original point pair feature. (a) Front view; (b) side view

    Fig. 1. Schematic of ambiguity of original point pair feature. (a) Front view; (b) side view

    Download full size
    Visible constraint between viewpoints. (a) Schematic of visible constraint between two points; (b) 3D model; (c) 2.5D scene

    Fig. 2. Visible constraint between viewpoints. (a) Schematic of visible constraint between two points; (b) 3D model; (c) 2.5D scene

    Download full size
    Viewpoint visibility constraint. (a) Location of point p; (b) variations in distribution and number of points satisfying visible constraint with point p with θ

    Fig. 3. Viewpoint visibility constraint. (a) Location of point p; (b) variations in distribution and number of points satisfying visible constraint with point p with θ

    Download full size
    Flow chart of 3D object recognition based on enhanced point pair feature

    Fig. 4. Flow chart of 3D object recognition based on enhanced point pair feature

    Download full size
    Dataset collected in practice. (a) Glass box and mouse model; (b) dataset R1; (c) dataset R2; (d) dataset R3

    Fig. 5. Dataset collected in practice. (a) Glass box and mouse model; (b) dataset R1; (c) dataset R2; (d) dataset R3

    Download full size
    Five models and two sample scenes of UWA dataset

    Fig. 6. Five models and two sample scenes of UWA dataset

    Download full size
    Recognition results of eight scenes on R1 dataset

    Fig. 7. Recognition results of eight scenes on R1 dataset

    Download full size
    Time cost comparison between EPPF and original PPF based methods on R1 dataset

    Fig. 8. Time cost comparison between EPPF and original PPF based methods on R1 dataset

    Download full size
    Recognition results of four scenes S1-S4 on R2 dataset

    Fig. 9. Recognition results of four scenes S1-S4 on R2 dataset

    Download full size
    Time cost comparison on R2 dataset. (a) EPPF; (b) original PPF

    Fig. 10. Time cost comparison on R2 dataset. (a) EPPF; (b) original PPF

    Download full size
    Recognition results of five scenes S1-S5 on R3 dataset

    Fig. 11. Recognition results of five scenes S1-S5 on R3 dataset

    Download full size
    Time cost comparison on R3 dataset. (a) EPPF; (b) original PPF

    Fig. 12. Time cost comparison on R3 dataset. (a) EPPF; (b) original PPF

    Download full size

    • Table 1. Summary of basic information of six target models

      View table

      Table 1. Summary of basic information of six target models

      ModelPoint numberOriginal PPFEnhanced PPF
      NumberTime cost /sNumberTime cost /s
      Chef33511122585085.10555742234.70
      Chicken2643698280644.00343495818.90
      Para2507628254243.20311446817.50
      T-rex2337545923235.90271320214.80
      Glass Box113412848223.946386221.87
      Mouse135818428066.169165922.96

    • Table 2. Recognition results of proposed algorithm on whole UWA dataset

      View table

      Table 2. Recognition results of proposed algorithm on whole UWA dataset

      ModelCorrect number/total numberFailed scene numberOcclusion of the targets /%
      Chef49/504391.30
      Chicken45/486,26,3289.70,86.50,89.50
      Parasaurolophus40/457,10,38,41,5086.40,91.40,89.00,87.00,83.90
      T-rex41/454,10,34,4884.00,80.20,83.80,77.30
      Average175/188 (93.1%)--

    • Table 3. Comparison of two algorithms on the UWA dataset (occlusion of targets is less than 84%) in terms of recognition rate

      View table

      Table 3. Comparison of two algorithms on the UWA dataset (occlusion of targets is less than 84%) in terms of recognition rate

      AlgorithmRecognition rate /%Time cost for one object /s
      Proposed97.610
      PPF in Ref. [8]97.085
    Tools

    Get Citation

    Copy Citation Text

    Rongrong Lu, Feng Zhu, Qingxiao Wu, Foji Chen, Yunge Cui, Yanzi Kong. Three-Dimensional Object Recognition Based on Enhanced Point Pair Features[J]. Acta Optica Sinica, 2019, 39(8): 0815006

    Download Citation

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

    Category: Machine Vision

    Received: Mar. 5, 2019

    Accepted: May. 5, 2019

    Published Online: Aug. 7, 2019

    The Author Email: Lu Rongrong (lurongrong@sia.cn), Zhu Feng (fzhu@sia.cn)

    DOI:10.3788/AOS201939.0815006

    Topics

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