《数字图像处理 Digital Image Processing》课件 Image Restoration.ppt

Deriving the blurring model Now our objective is to find the blurring function H(u,v). Recall two-dimensional Fourier transform formula Thus the Fourier transform of g(x,y) can be solved: Lecture 13: Chapter 5 Image Restoration Reversing the order of integration for G(u,v), we get the following form: Using the translation property of Fourier transform, we arrive: Deriving the blurring model Lecture 13: Chapter 5 Image Restoration Deriving the blurring model con’t Now it is clear that the integral part happens to be the blurring function H(u,v). So, G(u,v) is now expressed in the familiar form Lecture 13: Chapter 5 Image Restoration Now, to be more specific, the image in question undergoes uniform linear motion in the x-direction only, at a rate given by x0(t) = at/T. solving for H(u,v) yields: It is noted that H vanishes at values of u = n/a, where n is an integer. If y-component varies as well, according to y0 = bt/T, then H(u,v) becomes: Deriving the blurring model con’t Lecture 13: Chapter 5 Image Restoration Example to show image blurring due to motion Fig.b) is obtained by computing the Fourier transform of a), multiplying the transform by H(u,v), and taking the inverse transform Lecture 13: Chapter 5 Image Restoration Simulate motion blur using Matlab I = imread(Fig5.26(a).jpg); imshow(I); title(original image); PSF = fspecial(motion,100,135); MotionBlur = imfilter(I,PSF,circular,conv); figure, imshow(MotionBlur); title(Motion Blurred Image); Lecture 13: Chapter 5 Image Restoration Definition of Inverse filtering (逆滤波) Recall the image degradation model: If we divide G(u,v) by H(u,v) to get an estimate of F(u,v), then we get: This is called direct inverse filtering. Lecture 13: Chapter 5 Image Restoration Problems: 1) F(u,v) is a random function whose Fourier transform is not known. 2) If degradation function H(u,v) has zero or very small values, then the ratio N(u,v)/H(u,v) could easily dominate the estimate F(u,v). Solutions: from chapter 4, we already kn