Remote Sensing Technology and Application, Volume. 40, Issue 1, 226(2025)

A Local Registration Model based on Multidirectional Phase Congruency for High Resolution Optical Images

Wenhao AI and Xinghua LI*
Author Affiliations
  • School of Remote Sensing and Information Engineering, Wuhan University, Wuhan430079, China
  • show less
    Figures & Tables(13)
    The flow chart of the proposed method
    2D Log-Gabor filter spatial domain shape
    Feature matching results of mountain region images
    Feature matching results of agricultural region images
    Feature matching results of urban region images
    Feature matching results of mixed region images
    Registration results of test 1
    Registration results of test 2
    Registration results of test 3
    • Table 1. Experimental data information of feature matching

      View table
      View in Article

      Table 1. Experimental data information of feature matching

      影像标签成像时间分辨率/m影像类型影像尺寸数据特点
      山地GF12020.12.228多光谱1 003×1 027×4以山区为主,局部影像为平坦的城镇区域
      GF12020.04.12
      农区GF22020.03.051全色1 000×1 000×1纹理信息丰富且具有高重复性,存在非线性灰度映射
      GF22020.01.31
      城区GF62020.08.152全色1 412×1 502×1细部信息丰富并伴随有云雾、阴影遮挡
      GF12020.03.14
      混合区域GF12020.11.118多光谱1 194×1 243×4包括山区及城区
      GF12020.03.14
    • Table 2. The quantitative evaluation of feature matching

      View table
      View in Article

      Table 2. The quantitative evaluation of feature matching

      红色:最优值

      蓝色:次优值

      山区农区城区混合
      CMQCMRCMQCMRCMQCMRCMQCMR
      SIFT180.751300.8760.431210.88
      SURF60.752050.86250.49640.80
      本文提出算法400.741510.87790.983481.00
      HOPC390.701320.90750.672430.92
      LSS130.72470.90120.921120.99
    • Table 3. Experimental data information of image registration

      View table
      View in Article

      Table 3. Experimental data information of image registration

      影像标签成像时间分辨率/m影像类型影像尺寸数据特点
      测试1GF-72020.04.274多光谱6 842×7 428×4重叠区包含城区、农田、丘陵及水体,影像尺度存在差异,不存在明显的辐射畸变
      GF-12020.03.1484 803×5 629×4
      测试2GF-22020.05.184多光谱6 108×7 168×4重叠区为城区及郊区,局部山区地形,影像灰度具有较高相似性
      GF-72020.04.276 842×7 428×4
      测试3GF-12020.11.072全色19 191×22 036×1重叠区尺寸大,分辨率高,多为小尺寸地物,纹理特征混乱,存在非线性灰度映射
      GF-12020.05.1919 267×22 234×1
    • Table 4. The RMSE evaluation of image registration

      View table
      View in Article

      Table 4. The RMSE evaluation of image registration

      原始影像GPTMPMTM本研究算法
      测试111.690.911.11.130.78
      测试210.290.831.011.030.75
      测试312.552.162.282.421.64
    Tools

    Get Citation

    Copy Citation Text

    Wenhao AI, Xinghua LI. A Local Registration Model based on Multidirectional Phase Congruency for High Resolution Optical Images[J]. Remote Sensing Technology and Application, 2025, 40(1): 226

    Download Citation

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

    Category:

    Received: Mar. 20, 2023

    Accepted: --

    Published Online: May. 22, 2025

    The Author Email: Xinghua LI (lixinghua5540@whu.edu.cn)

    DOI:10.11873/j.issn.1004-0323.2025.1.0226

    Topics