With the development of neural networks in the field of image processing, deblurring is one of the important tasks of image processing, more and more excellent models have appeared one after another. We propose a classification-based strategy that assumes and verifies that the blurred images can be divided into different categories, and then feeds the blur images to the corresponding deblur models after classification. Our classification-based method can generally improve the performance of the current deblur models. After experimental verification, for the same dataset, our method improves PSNR from 30.26 to 31.15 and SSIM from 0.9342 to 0.9374.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Improving SRN-Deblur Performance Based on Blur Direction-based Classification


    Contributors:
    Xu, Xinyi (author) / Lu, Yue (author) / Xu, Feiran (author) / Zhao, Ming (author)


    Publication date :

    2022-10-12


    Size :

    1832354 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Learning Good Regions to Deblur Images

    Hu, Z. / Yang, M. H. | British Library Online Contents | 2015


    Blind nonuniform deblur under projection motion path

    Zhang, X. / Sun, F. | British Library Online Contents | 2013


    Comparison of Motion Deblur Algorithms and real world deployment

    Sehuon, Sebastian / Diepold, Klaus | AIAA | 2006


    Blur identification based on kurtosis minimization

    Dalong Li, / Mersereau, R.M. / Simske, S. | IEEE | 2005


    Blur Identification Based on Kurtosis Minimization

    Li, D. / Mersereau, R. M. / Simske, S. | British Library Conference Proceedings | 2005