自然图像盲复原的模糊参数估计：线性运动和失焦.pdfVIP

466 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 23, NO. 1, JANUARY 2014 Parametric Blur Estimation for Blind Restoration of Natural Images: Linear Motion and Out-of-Focus João P. Oliveira, Member, IEEE , Mário A. T. Figueiredo, Fellow, IEEE , and José M. Bioucas-Dias, Member, IEEE Abstract — This paper presents a new method to estimate Fourier domain), BID can be seen as the problem of recovering the parameters of two types of blurs, linear uniform motion two functions from their product; a clearly hopeless goal, in (approximated by a line characterized by angle and length) and the absence of strong assumptions or prior knowledge about out-of-focus (modeled as a uniform disk characterized by its radius), for blind restoration of natural images. The method is the underlying image and blur. Assumptions about the blur based on the spectrum of the blurred images and is supported PSF have included positiveness, known shape (e.g., Gaussian on a weak assumption, which is valid for the most natural blur), smoothness, symmetry, or known finite support [6]. images: the power-spectrum is approximately isotropic and has There are two main alternative approaches to BID: (i) a power-law decay with the spatial frequency. We introduce simultaneously estimate the image and the blur [1], [2], [9]; two modifications to the radon transform, which allow the identification of the blur spectrum pattern of the two types of (ii) obtain a blur estimate from the observed image and then blurs above mentioned. The blur parameters are identified by