数字图像处理实验（13）：PROJECT 05-04，Parametric Wiener Filter ...

实验要求：

Objective:
To understand the high performance of the parametric Wiener Filter in image restoration when there are additive noise after the image degradation.
Main requirements:
Ability of programming with C, C++, or Matlab.
Instruction manual:
(a) Implement a blurring filter as in Eq. (5.6-11).
(b) Blur image 5.26(a) in the +45o direction using T = 1, as in Fig. 5.26(b).
(c) Add Gaussian noise of 0 mean and variance of 10 pixels to the blurred image.
(d) Restore the image using the parametric Wiener filter given in Eq. (5.8-3).

本实验属于图像复原技术，使用参数维纳滤波进行图像复原。实验中向图像添加了高斯噪声和运动模糊，最后用参数维纳滤波器复原图像。

%
close all;
clc;
clear all;

% 读取图像
img = imread('Fig5.26(a).jpg');
img = im2double(img);
figure;
subplot(2,3,1);
imshow(img);
title('original image');

% 模糊图像
PSF = fspecial('motion', 30, 45);
img1 = imfilter(img, PSF, 'conv', 'circular');
subplot(2,3,2);
imshow(img1);
title('filtered image');

% 添加高斯噪声
noise_var = 0.001;
img2 = imnoise(img1, 'gaussian', 0, noise_var);
subplot(2,3,3);
imshow(img2);
title('add gaussian noise');

% 参数维纳滤波，NSR直接给0
% Specifying 0 for the NSR is equivalent to creating an ideal inverse filter.
% img3 = deconvwnr(img2, PSF, 0.012);
img3 = deconvwnr(img2, PSF, 0.0);
subplot(2,2,3);
imshow(img3);
title('Restoration of Blurred, Noisy Image Using NSR = 0');

% 参数维纳滤波，计算方差
% img = double(img);
estimated_NSR = noise_var / var(img(:));
img4 = deconvwnr(img2, PSF, estimated_NSR);
subplot(2,2,4);
imshow(img4);
title('Restoration of Blurred, Noisy Image Using Estimated NSR');

实验结果：

上面一行的图像分别是原始图像，模糊后的图像，以及添加高斯噪声后的图像；
下面一行的图像分别是调用维纳滤波器的两种情况，一个是不给参数，默认直接给0，另一个是使用方差计算参数后调用维纳滤波器得到的正确滤波结果。

转载于:https://www.cnblogs.com/xuhongbin/p/7134157.html