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

Laser & Optoelectronics Progress, Volume. 57, Issue 16, 161503(2020)

Fully Automatic Registration of Remote Sensing Images Fusion of Point and Line Complementary Features

Guobiao Yao1、*, Xiaocheng Man1, Chuanhui Zhang1, Qingqing Fu2, Guoqiang Zheng1, and Bing Li1
Author Affiliations
  • 1School of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan, Shandong 250101, China
  • 2School of Mechanical and Electronic Engineering, Shandong Jianzhu University, Jinan, Shandong 250101, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (14)
    Equations (0)
    References (17)
    Cited By (1)
    Tools
    Paper Information
    Topics
    Figures & Tables(14)
    Schematic of matching process of proposed algorithm

    Fig. 1. Schematic of matching process of proposed algorithm

    Download full size
    SIFT matching point matching refinement result

    Fig. 2. SIFT matching point matching refinement result

    Download full size
    Process of matching straight lines with homography geometric constraint

    Fig. 3. Process of matching straight lines with homography geometric constraint

    Download full size
    Process of matching straight lines with slope constraint

    Fig. 4. Process of matching straight lines with slope constraint

    Download full size
    Descriptor construction

    Fig. 5. Descriptor construction

    Download full size
    4 sets of images in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Fig. 6. 4 sets of images in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Download full size
    Matching results of SIFT algorithm in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Fig. 7. Matching results of SIFT algorithm in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Download full size
    Matching results of proposed algorithm in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Fig. 8. Matching results of proposed algorithm in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Download full size
    Extraction results of straight line features. (a) Original image; (b) LSD algorithm; (c) proposed algorithm

    Fig. 9. Extraction results of straight line features. (a) Original image; (b) LSD algorithm; (c) proposed algorithm

    Download full size
    Line feature matching results of proposed algorithm in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Fig. 10. Line feature matching results of proposed algorithm in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Download full size
    Complementary fusion of dot-line features in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Fig. 11. Complementary fusion of dot-line features in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Download full size
    Registration results of proposed algorithm in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Fig. 12. Registration results of proposed algorithm in different situations. (a) Rotation change; (b) scale change; (c) lack of texture; (d) light and dark change

    Download full size

    • Table 1. Comparison of matching results between SIFT algorithm and proposed algorithm

      View table

      Table 1. Comparison of matching results between SIFT algorithm and proposed algorithm

      ImageSIFT algorithmImageProposed algorithm
      SumCorrectCorrectrate /%εRMSE /pixelt /sSumCorrectCorrectrate /%εRMSE /pixelt /s
      7(a)110110019029.477.28(a)103910391000.299.3
      7(b)2402128867.683.48(b)229226990.295.5
      7(c)7206148520.502.88(c)7007001000.305.4
      7(d)5645089029.807.68(d)538532990.309.4

    • Table 2. Comparison of matching results of line features

      View table

      Table 2. Comparison of matching results of line features

      ImageLine featureAlgorithmSumCorrectRate /%
      Ref. [11]40534083
      10(a)1654/1333Ref. [12]46542290
      Proposed64362196
      Ref. [11]1126658
      10(c)1015/1139Ref. [12]23522495
      Proposed28628198
      Ref. [11]18116188
      10(c)576/676Ref. [12]21720192
      Proposed375375100
      Ref. [11]21513964
      10(d)1044/664Ref. [12]33230190
      Proposed50047896
    Tools

    Get Citation

    Copy Citation Text

    Guobiao Yao, Xiaocheng Man, Chuanhui Zhang, Qingqing Fu, Guoqiang Zheng, Bing Li. Fully Automatic Registration of Remote Sensing Images Fusion of Point and Line Complementary Features[J]. Laser & Optoelectronics Progress, 2020, 57(16): 161503

    Download Citation

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

    Category: Machine Vision

    Received: Dec. 4, 2019

    Accepted: Jan. 16, 2020

    Published Online: Aug. 5, 2020

    The Author Email: Guobiao Yao (yao7837005@163.com)

    DOI:10.3788/LOP57.161503

    Topics

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