Chinese Journal of Liquid Crystals and Displays, Volume. 37, Issue 4, 483(2022)

Adaptive low overlap multispectral image fast mosaic algorithm based on phase correlation enhancement

Ting-ting JIA1, Hui-qin WANG1、*, Ke WANG1, Zhan WANG2, Gang ZHEN2, and Yuan LI3
Author Affiliations
  • 1College of Information and Control Engineering,Xi’an University of Architecture and Technology,Xi’an 710055,China
  • 2Shanxi Provincial Institute of Cultural Relics Protection,Xi’an 710075,China
  • 3Xi’an Museum,Xi’an 710054,China
  • show less
    Figures & Tables(12)
    Pulse diagram of phase correlation function of image.(a)Correctly match the pulse diagram;(b)Error matching pulse diagram.
    Schematic diagram of image segmentation
    Flow chart of adaptive segmentation algorithm
    Schematic diagram of multispectral image A1 and A2 splicing
    Comparison of stitching effect between traditional phase correlation method and this algorithm.(a)Image to be spliced;(b)Traditional phase correlation pulse diagram;(c)Pulse diagram in this paper;(d)Splicing result diagram of traditional phase correlation method;(e)Stitching results of this algorithm shown in the figure.
    Mosaic results of spectral images of different murals
    Comparison of splicing effects of different algorithms
    Information entropy change diagram of 16 channels of A1 and A2 images
    16 channel spectral image mosaic effect
    • Table 1. Mosaic results of five groups of mural spectral images

      View table
      View in Article

      Table 1. Mosaic results of five groups of mural spectral images

      ImageImage size/pixelOverlap rate/%Segmentation timesΔxΔytime/s
      A1 A21 363×96511.451761100.412
      B1 B21 382×99755771060.405
      C1 C21 356×9987.810116850.793
      D1 D21 378×1 00020.310300680.817
      E1 E21 361×94715.85219160.415
    • Table 2. Comparison of stitching results of different algorithms

      View table
      View in Article

      Table 2. Comparison of stitching results of different algorithms

      WavelengthAlgorithmEvaluation index
      SSIMPSNRRMSETime/s
      520 nmHarris0.814 628.245 19.869 21.325 4
      SIFT0.966 236.248 92.345 71.742 5
      FAST-SURF0.943 232.854 15.421 30.972 4
      BRISK-ORB0.915 431.125 87.235 40.391 5
      Ours0.987 842.563 21.754 20.361 8
      740 nmHarris0.945 331.254 65.426 51.425 6
      SIFT0.985 641.265 41.863 41.845 9
      FAST-SURF0.981 540.786 92.045 81.043 6
      BRISK-ORB0.952 332.865 75.102 30.405 3
      Ours0.990 244.882 31.452 30.395 2
      900 nmHarris0.808 028.453 69.632 11.386 5
      SIFT0.956 433.245 13.852 41.802 3
      FAST-SURF0.952 132.742 53.245 71.005 0
      BRISK-ORB0.892 529.213 58.334 30.398 7
      Ours0.972 439.564 22.564 20.372 4
    • Table 3. Comparison of 16-channel splicing results

      View table
      View in Article

      Table 3. Comparison of 16-channel splicing results

      Evaluation indexAlgorithm
      HarrisSIFTFAST-SURFBRISK-ORBOurs
      IE/bitMax7.439 67.545 47.529 67.486 47.625 4
      Min2.214 72.352 72.295 72.258 42.601 2
      Mean6.027 56.175 46.135 46.100 66.254 1
      AGMax8.863 29.653 69.542 59.011 510.235 1
      Min1.145 21.615 41.496 70.215 31.738 7
      Mean4.845 54.956 24.945 14.825 05.008 1
      Time/sSum7.8529.5635.8674.3513.424
    Tools

    Get Citation

    Copy Citation Text

    Ting-ting JIA, Hui-qin WANG, Ke WANG, Zhan WANG, Gang ZHEN, Yuan LI. Adaptive low overlap multispectral image fast mosaic algorithm based on phase correlation enhancement[J]. Chinese Journal of Liquid Crystals and Displays, 2022, 37(4): 483

    Download Citation

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

    Category:

    Received: Nov. 18, 2021

    Accepted: --

    Published Online: Jun. 20, 2022

    The Author Email: Hui-qin WANG (hqwang@xauat.edu.cn)

    DOI:10.37188/CJLCD.2021-0294

    Topics