Turbulent disturbance brings about the spatially varying blur in captured images. Turbulent disturbance brings about the spatially varying blur in captured images. In this paper, we proposed a deep learning method to deal with this problem. And the way we approximate the kernel is closer to the real turbulent circumstance. Experiments on both synthetic and real images show that our proposed method can effectively mitigate the turbulent blur from a single image, quantitatively and qualitatively outperforms the single Preprint submitted to Neurocomputing August 23, 2023frame based methods as well as yields comparable results to multi-frame methods.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Deep Kernel Estimation for Turbulent Blur Correction


    Contributors:


    Publication date :

    2023-10-11


    Size :

    3047616 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Blur correction device and imaging device

    FURUKAWA MASAHIDE / SUGIYAMA YOSHINORI | European Patent Office | 2021

    Free access


    Range Estimation from Camera Blur by Regularized Adaptive Identification

    Holeva, L. F. / SPIE | British Library Conference Proceedings | 1993



    Double Regularized Bayesian Estimation for Blur Identification in Video Sequences

    Zheng, H. / Hellwich, O. | British Library Conference Proceedings | 2006