Acta Optica Sinica, Volume. 39, Issue 7, 0715003(2019)

Image Super Resolution Based on Depth Jumping Cascade

Kunpeng Yuan* and Zhihong Xi
Author Affiliations
  • College of Information and Communication Engineering, Harbin Engineering University, Harbin, Heilongjiang 150001, China
  • show less
    Figures & Tables(15)
    Diagram of SRCNN network structure
    Diagram of VDSR network structure
    Network structural diagram of proposed algorithm
    Jumping cascade block
    Visualization results of picture butterfly after 4-time magnification. (a) Low-level feature; (b) hopping cascade feature; (c) reconstructing feature
    Relationship between average PSNR of different layers with number of iterations under Set5 test set
    Relationship between iteration times and average PSNR of different network structures under Set5 test set
    Relationship between average PSNR of different activation functions with number of iterations under Set5 test set
    Relationship between average PSNR of different filter numbers with number of iterations under Set5 test set
    Relationship between run time and average PSNR of different methods under Set5 test set
    Comparison of butterfly images obtained by different algorithms
    Comparison of PPT obtained by different algorithms
    Comparison of man images obtained by different algorithms
    • Table 1. Average PSNR of test sets Set5, Set14, and BSD100 under different algorithms

      View table

      Table 1. Average PSNR of test sets Set5, Set14, and BSD100 under different algorithms

      DatasetScaleBicubicSRCNN[12]ESPCN[14]FSRCNN[13]VDSR[15]DCSR
      233.6836.1936.3836.4537.3437.70
      Set5330.4532.4632.7132.5933.4734.13
      428.4630.1530.2930.4230.7831.87
      230.2132.132.232.2132.8233.26
      Set14327.5128.9929.1229.1229.5129.97
      425.9827.2327.1727.4327.6228.27
      229.4330.8830.9331.2431.5131.81
      BSD100327.0828.0628.1628.2528.4328.79
      425.8426.6326.5926.8526.8727.28
    • Table 2. Average SSIM of test sets Set5,Set14, and BSD100 under different algorithms

      View table

      Table 2. Average SSIM of test sets Set5,Set14, and BSD100 under different algorithms

      DatasetScaleBicubicSRCNN[12]ESPCN[14]FSRCNN[13]VDSR[15]DCSR
      20.93060.95510.95680.95670.95800.9636
      Set530.86860.91100.91500.91220.91880.9321
      40.81020.86210.86290.86590.87500.8979
      20.86930.95760.95970.96340.91040.9644
      Set1430.77440.88360.88730.89230.82710.8981
      40.70230.82070.82270.82730.75920.8446
      20.84400.88010.88310.88670.89240.8964
      BSD10030.74010.77550.78110.78030.79280.7987
      40.66970.6920.69430.70090.71860.7203
    Tools

    Get Citation

    Copy Citation Text

    Kunpeng Yuan, Zhihong Xi. Image Super Resolution Based on Depth Jumping Cascade[J]. Acta Optica Sinica, 2019, 39(7): 0715003

    Download Citation

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

    Category: Machine Vision

    Received: Jan. 8, 2019

    Accepted: Mar. 22, 2019

    Published Online: Jul. 16, 2019

    The Author Email: Yuan Kunpeng (xizhihong@hrbeu.edu.cn)

    DOI:10.3788/AOS201939.0715003

    Topics