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. 38, Issue 10, 1010003(2018)

Joint Deep Denoising Prior for Image Blind Deblurring

Aiping Yang*, Jinbin Wang, Bingwang Yang, and Yuqing He
Author Affiliations
  • School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (12)
    Equations (0)
    References (0)
    Cited By (7)
    Tools
    Paper Information
    Topics
    Figures & Tables(12)
    Whole framework of algorithm

    Fig. 1. Whole framework of algorithm

    Download full size
    Pixel gray histograms of (a) clear image and (b) blurred image

    Fig. 2. Pixel gray histograms of (a) clear image and (b) blurred image

    Download full size
    Structure of denoising deep convolution neural network

    Fig. 3. Structure of denoising deep convolution neural network

    Download full size
    Texture layer and structure layer of image. (a) Original image; (b) texture layer; (c) structure layer

    Fig. 4. Texture layer and structure layer of image. (a) Original image; (b) texture layer; (c) structure layer

    Download full size
    Deblur results of different algorithms. (a) Blur images; (b) method in Ref. [7]; (c) method in Ref. [33]; (d) method in Ref. [16]; (e) proposed method

    Fig. 5. Deblur results of different algorithms. (a) Blur images; (b) method in Ref. [7]; (c) method in Ref. [33]; (d) method in Ref. [16]; (e) proposed method

    Download full size
    Clear images. (a) Boys; (b) bridge; (c) paint; (d) face

    Fig. 6. Clear images. (a) Boys; (b) bridge; (c) paint; (d) face

    Download full size
    Blur kernel estimation of different algorithms. (a) True blur kernel; (b) method in Ref. [16]; (c) method in Ref. [7]; (d) method in Ref. [32]; (e) method in Ref. [34]; (f) proposed method

    Fig. 7. Blur kernel estimation of different algorithms. (a) True blur kernel; (b) method in Ref. [16]; (c) method in Ref. [7]; (d) method in Ref. [32]; (e) method in Ref. [34]; (f) proposed method

    Download full size

    • Table 1. Comparison of performance and time complexity of denoising convolution neural networks

      View table

      Table 1. Comparison of performance and time complexity of denoising convolution neural networks

      DatasetPSNR /dBSSIMt /s
      ProposedmethodMethod inRef. [19]ProposedmethodMethod inRef. [19]ProposedmethodMethod inRef. [19]
      BSD6828.7429.220.80350.8278323606
      Classic529.6230.380.80420.83232262
      Set1229.7930.420.83810.86176580
      Set1429.3430.010.80990.83523719

    • Table 2. Iterative algorithm for image deblurring

      View table

      Table 2. Iterative algorithm for image deblurring

      input: blur image B and blur kernel k
      I ←B,β ←2λσrepeat: solve for u using Eq. (21)μ ←2λ repeat: solve for g using Eq. (22) solve for z using Eq. (23) solve for x using Eq. (17)μ ←2μ until μ>μmaxβ ←2βuntil β>βmaxoutput:internediate latent image x→I

    • Table 3. Iterative algorithm for blur kernel estimation

      View table

      Table 3. Iterative algorithm for blur kernel estimation

      input: blur image B
      initialize I and k with the results from the coarser level;for j=1→5 dosolve for IS using Eq. (11)solve for k using Eq. (23)solve x→I using iterative algorithm in table 1λ←maxλ/1.1,1×e-4end foroutput:blur kernel k and internediate latent image x

    • Table 4. PSNR results of different algorithmsdB

      View table

      Table 4. PSNR results of different algorithmsdB

      BlurImageMethod inRef. [16]Method inRef. [7]Method Ref. [32]Method inRef. [34]Proposedmethod
      1Boys25.7124.7427.1125.3030.00
      Bridge27.7424.7224.4918.9727.72
      Paint26.4623.4025.1023.0029.99
      Face24.1425.9026.3626.3829.09
      2Boys31.2125.4228.6722.7933.23
      Bridge28.3126.4128.7920.1629.40
      Paint30.9925.6228.5724.7931.99
      Face27.8026.0528.8223.6430.31
      3Boys27.0324.2225.8718.9030.84
      Bridge26.2422.4923.4821.7728.11
      Paint22.3823.7625.8618.5228.46
      Face24.1126.6025.4925.9525.83
      4Boys30.7526.2429.0323.0832.38
      Bridge23.3925.1826.2225.6428.33
      Paint26.8925.5225.9833.0530.46
      Face25.6423.7124.7922.5829.52

    • Table 5. Time complexity of different algorithmss

      View table

      Table 5. Time complexity of different algorithmss

      ImageImage size /(pixel×pixel)Blur kernelsize /(pixel×pixel)Method inRef. [16]Method inRef. [33]Method inRef. [34]Proposedmethod
      ECCV123×12627×2710119916830
      Roma593×41735×358192938792138
      Cartoon612×44219×197262880249168
      Flower900×89635×35290986431921358
    Tools

    Get Citation

    Copy Citation Text

    Aiping Yang, Jinbin Wang, Bingwang Yang, Yuqing He. Joint Deep Denoising Prior for Image Blind Deblurring[J]. Acta Optica Sinica, 2018, 38(10): 1010003

    Download Citation

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

    Category: Image Processing

    Received: Mar. 19, 2018

    Accepted: May. 21, 2018

    Published Online: May. 9, 2019

    The Author Email:

    DOI:10.3788/AOS201838.1010003

    Topics

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