-
Notifications
You must be signed in to change notification settings - Fork 172
/
Demo_deblur_real_application.m
166 lines (134 loc) · 4.76 KB
/
Demo_deblur_real_application.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
%==========================================================================
% This is the testing code of IRCNN for image deblurring with estimated
% kernel by other blind deblurring methods.
%
% There are two important parameters to tune:
% (1) image noise level of blurred image: Isigma and
% (2) noise level of the last denoiser: Msigma.
%
% @inproceedings{zhang2017learning,
% title={Learning Deep CNN Denoiser Prior for Image Restoration},
% author={Zhang, Kai and Zuo, Wangmeng and Gu, Shuhang and Zhang, Lei},
% booktitle={IEEE Conference on Computer Vision and Pattern Recognition},
% year={2017}
% }
%
% If you have any question, please feel free to contact with <Kai Zhang ([email protected])>.
%
%
% by Kai Zhang (1/2018)
%==========================================================================
clear; clc;
addpath('utilities');
imageSets = {'Deblur_set1','Deblur_set2','Deblur_set3'}; % testing dataset
setTest = imageSets(2); % select the dataset
useGPU = 1;
folderTest = 'testsets';
folderResult = 'results';
folderModel = 'models';
if ~exist(folderResult,'file')
mkdir(folderResult);
end
setTestCur = cell2mat(setTest(1));
disp('--------------------------------------------');
disp(['----',setTestCur,'-----Image Debluring-----']);
disp('--------------------------------------------');
folderTestCur = fullfile(folderTest,setTestCur);
% folder to store results
folderResultCur = fullfile(folderResult, ['Deblur_',setTestCur]);
if ~exist(folderResultCur,'file')
mkdir(folderResultCur);
end
%% read blurred image and its estimated kernel
% blurred image
Iname = 'im01_ker01';
y = im2single(imread(fullfile(folderTestCur,[Iname,'.png'])));
% estimated kernel
%k = imread(fullfile(folderTestCur,[Iname,'_kernel.png']));
k = imread(fullfile(folderTestCur,[Iname,'_out_kernel.png']));
if size(k,3)==3
k = rgb2gray(k);
end
k = im2single(k);
k = k./(sum(k(:)));
%% -------------------important!------------------
% Parameter settings of IRCNN
% (1) image noise level of blurred image: Isigma
Isigma = 5/255; % ****** from interval [1/255, 20/255] ******; e.g., 1/255, 2.55/255, 7/255, 11/255
% (2) noise level of the last denoiser: Msigma
Msigma = 5; % ****** from {1 3 5 7 9 11 13 15} ******
%--------------------------------------------------------
[a1,b1,~] = size(y);
%% handle boundary
boundary_handle = 'case2';
switch boundary_handle
case {'case1'} % option (1), edgetaper to better handle circular boundary conditions, (matlab2015b)
% k(k==0) = 1e-10; % uncomment this for matlab 2016--2018?
ks = floor((size(k) - 1)/2);
y = padarray(y, ks, 'replicate', 'both');
for a=1:4
y = edgetaper(y, k);
end
case {'case2'} % option (2)
H = size(y,1); W = size(y,2);
y = wrap_boundary_liu(y, opt_fft_size([H W]+size(k)-1));
end
[w,h,c] = size(y);
V = psf2otf(k,[w,h]);
denominator = abs(V).^2;
if c>1
denominator = repmat(denominator,[1,1,c]);
V = repmat(V,[1,1,c]);
end
upperleft = conj(V).*fft2(y);
% load denoisers
if c==1
load(fullfile(folderModel,'modelgray.mat'));
elseif c==3
load(fullfile(folderModel,'modelcolor.mat'));
end
totalIter = 30; % default 30
lamda = (Isigma^2)/3; % default 3, ****** from {1 2 3 4} ******
modelSigma1 = 49; % default 49
modelSigmaS = logspace(log10(modelSigma1),log10(Msigma),totalIter);
rho = Isigma^2/((modelSigma1/255)^2);
ns = min(25,max(ceil(modelSigmaS/2),1));
ns = [ns(1)-1,ns];
z = single(y);
if useGPU
z = gpuArray(z);
upperleft = gpuArray(upperleft);
denominator = gpuArray(denominator);
end
for itern = 1:totalIter
% step 1
rho = lamda*255^2/(modelSigmaS(itern)^2);
z = real(ifft2((upperleft + rho*fft2(z))./(denominator + rho)));
if ns(itern+1)~=ns(itern)
[net] = loadmodel(modelSigmaS(itern),CNNdenoiser);
net = vl_simplenn_tidy(net);
if useGPU
net = vl_simplenn_move(net, 'gpu');
end
end
% step 2
res = vl_simplenn(net, z,[],[],'conserveMemory',true,'mode','test');
residual = res(end).x;
z = z - residual;
% imshow(z)
% title(int2str(itern))
% drawnow;
end
if useGPU
output = im2uint8(gather(z));
end
switch boundary_handle
case {'case1'} % option (1)
output = center_crop(output,a1,b1);
y = center_crop(y,a1,b1);
case {'case2'} % option (2)
output = output(1:a1,1:b1,:);
y = y(1:a1,1:b1,:);
end
imshow(cat(2,im2uint8(y),output));
imwrite(output,fullfile(folderResultCur,[Iname,'_ircnn_Isigma_',int2str(Isigma*255),'_Msigma_',int2str(Msigma),'.png']));