A Fast Combined Algorithm for Blind Image Restoration

来源 :第八届工业与应用数学国际大会 | 被引量 : 0次 | 上传用户:ddr133
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
  The blind image restoration problem is not an easy problem to solve for its complexity and difficulty.In this paper,we propose a new algorithm to simultaneously recover the blurring kernel and the original image,which combines split Bregman technique,fast Fourier transform and spectral decomposition technology to accelerate the computation of blind image restoration problem.
其他文献
  Consensus control of multi-agent systems has received compelling attention.Due to the large size of agents,the spatial distribution of actuators,limited sen
会议
  This talk investigates a new class of linear multi-agent systems,in which nodes are coupled by dynamic edges in the sense that each edge has a dynamic syste
会议
  The dynamics of a complex network is generally very complicated due to the self-dynamics of the node and their interactions.Many existing conditions for rea
会议
  In the investigation of multi-agent systems(MAS),a central issue is to understand how local interactions between agents lead to collective behavior of the s
会议
  After reviewing experiments involving thin films of suspensions and solutions,we propose ways to construct gradient dynamical models for liquid films of sus
会议
  The bulk behaviour of super-spreading surfactants is well modelled by the thin-film equations.However,an essential feature of the spreading mechanism,surfac
会议
  Dewetting from a liquid substrate is in many respects very similar to dewetting from a solid substrate,except that it has a much richer structure.We present
会议
  Thin-film type equations have important applications,e.g.in coating flows,food processing,solar cells.Recently research has focussed on the inclusion of mor
会议
  ADMM is a popular first order method.However,it suffers from solving linear subproblems in various applications.We give a preconditioned ADMM method by writ
会议
  Many imaging tasks amount to solve inverse problems.They are typically solved by minimizing an objective that accounts for the models of the recording devic
会议