/
Main_application_validation.m
362 lines (318 loc) · 15 KB
/
Main_application_validation.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
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% Main function to use distortion functions
%% Copyright (c) 2023, AYMAN BEGHDADI
%% All rights reserved.
%% Author: Ayman Beghdadi
%% Email: aymanaymar.beghdadi@univ-evry.fr
%% Date: January 2023
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% Function main_distortion:
% Call the different functions to generate distortions on the train set of
% the MS-COCO 2017 dataset
%
clear all;
clc;
addpath('./Distortions functions');
addpath('./Defocus local functions');
addpath('./Motion local functions');
addpath('./Other functions');
%% Path declarations for COCO dataset
% Dataset image for the training
imgtrain_path='./val2017';
%Add path
addpath(imgtrain_path);
%% Path declaration for the output distortion directory
% Dataset of the distorted image for the training
outputFolder='./val2017_distorted/';
% Add path
addpath(outputFolder);
%% Create the distortion directories
% Specified the path to your annotations files from train, val, or test
% folders
path_annotation =('./annotations_unpacked_valfull2017\matFiles');
addpath(path_annotation);
%% Specific directories for haze and rain sources
hazeFolder=('./weather_image\haze/');
addpath(hazeFolder);
rainFolder=('./weather_image\rain/');
addpath(rainFolder);
haze_img = "./weather_image\haze\fog_";
rain_img = "rain_";
rain_img2 = "rain2_";
%% Scene annoatations
load("scene_annotation_val.mat","Scene");
load("weather_val1.mat","Rain","Haze");
%% Image names
fullFileName = fullfile(pwd, 'ImageList_val.txt');
Image_name_list = readtable(fullFileName);
%% Additional path
Depth_location="./Depth_validation/";
Depth_end="-dpt_beit_large_512.png";
Depth_endss="-dpt_swin2_base_384.png";
Depth_ends="-dpt_swin2_large_384.png";
Image_location="./val2017/";
Image_end=".jpg";
%% Define all parameters for the various distortion functions (excepted name
% ,imgin and outputFolder)
contrast = [0.3 0.7;0.2 0.8;0.2 1.0;0.3 0.8;0.2 0.9; 0.3 0.9;...
0.3 1.0;0.35 0.9;0.35 1.0;0.4 0.9];
Global_distortion_choice = Select_global_distortion(Scene);
save("Global_choice_val_test.mat","Global_distortion_choice");
% load("Global_choice.mat","Global_distortion_choice");
%% Loop distortion performing
% breaks down annotation extraction into 6 steps to not overload MATLAB capabilities
% for id=1:size(Scene,1)
tic
for id=1:length(Scene)
% Get the image name
Image_name = char(table2cell(Image_name_list(id,:)));
Name = Image_name(1:end-4);
% Get the scene annotation
Scene_annotation = Scene(id);
% Path to the .mat data file
Data_ = [path_annotation '/' Name '.mat'];
% Give the image and depth complet name
Idepth = Depth_location+Name+Depth_end;
Dimg=Image_location+Name+Image_end;
distortion = "none";
% Check if the depth image exists
if (isfile(Idepth)==0)
Idepth = Depth_location+Name+Depth_ends;
if (isfile(Idepth)==0)
Idepth = Depth_location+Name+Depth_endss;
end
end
if (isfile(Data_))
if((Rain(id,1)==0) && (Haze(id,1)==0) && (Scene_annotation.weather==1))
Scene_annotation.global=1;
end
if((Scene_annotation.global==0) && (Scene_annotation.motion==0) && (Scene_annotation.weather==0) && (Scene_annotation.defocus==0) && (Scene_annotation.perturbation==0) && (Scene_annotation.back==0))
Scene_annotation.global=1;
end
if((Rain(id,1)==0) && (Haze(id,1)==0) && (Scene_annotation.weather==1))
Scene_annotation.global=1;
end
D=imread(Dimg);
if(Scene_annotation.global==1)
if(Scene_annotation.motion==1 || Scene_annotation.defocus==1)
choice = randi([3,5],1);
% Check if noise distortion is selected and if so apply it
if(choice==3)
means = 0;
variance = randi([1 20],1)*0.001;
[distortion, D] = distortion_noise(D,Image_name,means,variance,outputFolder);
end
% Check if compression distortion is selected and if so apply it
if(choice==4)
compression = randi([10 50],1);
[distortion, D] = distortion_compression(D,Image_name,compression,outputFolder);
% distortion=distortion;
end
% Check if contrast distortion is selected and if so apply it
if(choice==5)
cn = randi([1 10],1);
[distortion, D] = Global_contraste(Image_name, D,contrast(cn,:),outputFolder);
end
else
% Check if global motion distortion is selected and if so apply it
if(Global_distortion_choice(id,1)==1)
angle = randi([1 360],1);
len = randi([1 15],1);
[distortion, D] = distortion_motion_blur(D,Image_name,len,angle,outputFolder);
end
% Check if global defocus distortion is selected and if so apply it
if(Global_distortion_choice(id,1)==2)
gaussian_size = 9;
gaussian_sigma = randi([1 20],1)*0.1;
[distortion, D] = distortion_defocus_blur(D,Image_name,gaussian_size,gaussian_sigma,outputFolder);
end
% Check if noise distortion is selected and if so apply it
if(Global_distortion_choice(id,1)==3)
means = 0;
variance = randi([1 20],1)*0.001;
[distortion, D] = distortion_noise(D,Image_name,means,variance,outputFolder);
end
% Check if compression distortion is selected and if so apply it
if(Global_distortion_choice(id,1)==4)
compression = randi([10 50],1);
[distortion, D] = distortion_compression(D,Image_name,compression,outputFolder);
end
% Check if contrast distortion is selected and if so apply it
if(Global_distortion_choice(id,1)==5)
cn = randi([1 10],1);
[distortion, D] = Global_contraste(Name, D,contrast(cn,:),outputFolder);
end
end
else
if(Scene_annotation.perturbation==1)
outputname = sprintf('%s',Image_name);
imwrite(D, [outputFolder outputname],'Quality',95);
distortion= "none";
end
if(Scene_annotation.back==1)
data=load(Data_);
cn=randi([1,10],1);
[distortion, D] = Local_Backlight(Name,D,data,contrast(cn,:),outputFolder);
end
if(Scene_annotation.motion==1)
if(isfile(Idepth))
data=load(Data_);
I=imread(Idepth);
if(Scene_annotation.type==1)
scene="outdoor";
else
scene="indoor";
end
[distortion, D] = Local_blur_motion(Image_name,I,D,data,outputFolder,scene);
else
angle = randi([1 360],1);
len = randi([1 15],1);
[distortion, D] = distortion_motion_blur(D,Image_name,len,angle,outputFolder);
end
end
if(Scene_annotation.defocus==1)
if(isfile(Idepth))
data=load(Data_);
I=imread(Idepth);
[distortion, D] = Local_defocus_blur(Image_name,I,D,data,outputFolder);
else
gaussian_size = 9;
gaussian_sigma = randi([1 20],1)*0.1;
[distortion, D] = distortion_defocus_blur(D,Image_name,gaussian_size,gaussian_sigma,outputFolder);
end
end
if(Scene_annotation.weather==1)
data=load(Data_);
% I=imread(Idepth);
if(((Rain(id,1)==0) && (Haze(id,1)==0) && (Scene_annotation.weather==1)) ||(isfile(Idepth)==0))
% I=imread(Idepth);
choice = randi([1,5],1);
if(choice==1)
angle = randi([1 360],1);
len = randi([1 15],1);
[distortion, D] = distortion_motion_blur(D,Image_name,len,angle,outputFolder);
end
% Check if global defocus distortion is selected and if so apply it
if(choice==2)
gaussian_size = 9;
gaussian_sigma = randi([1 20],1)*0.1;
[distortion, D] = distortion_defocus_blur(D,Image_name,gaussian_size,gaussian_sigma,outputFolder);
end
% Check if noise distortion is selected and if so apply it
if(choice==3)
means = 0;
variance = randi([1 20],1)*0.001;
[distortion, D] = distortion_noise(D,Image_name,means,variance,outputFolder);
end
% Check if compression distortion is selected and if so apply it
if(choice==4)
compression = randi([10 50],1);
[distortion, D] = distortion_compression(D,Image_name,compression,outputFolder);
end
% Check if contrast distortion is selected and if so apply it
if(choice==5)
cn = randi([1 10],1);
[distortion, D] = Global_contraste(Name, D,contrast(cn,:),outputFolder);
end
end
if(Haze(id,1)==1 && isfile(Idepth))
I=imread(Idepth);
if(Haze(id,2)==1 ||Haze(id,2)==2)
alpha = 0.01 * randi([65,95],1);
img = randi([477,547],1);
file = haze_img + num2str(img) +Image_end;
Haze_= imread(file);
[distortion, D] = Global_haze(Name,D,I,Haze_,alpha,outputFolder);
end
if(Haze(id,2)==3)
alpha = 0.01 * randi([65,95],1);
img = randi([2216,2276],1);
file = haze_img + num2str(img) +Image_end;
Haze_= imread(file);
[distortion, D] = Global_haze(Name,D,I,Haze_,alpha,outputFolder);
end
end
if(Rain(id,1)==1 && isfile(Idepth))
I=imread(Idepth);
if(Rain(id,2)==1)
r = randi([1,55],1);
img_rain="./weather_image\rain\left\rain\rain_";
file = img_rain + num2str(r) +Image_end;
Rain_= imread(file);
end
if(Rain(id,2)==2)
r = randi([1,55],1);
img_rain = "./weather_image\rain\middle\rain2\rain2_";
file = img_rain + num2str(r) +Image_end;
Rain_= imread(file);
end
if(Rain(id,2)==3)
r = randi([1,55],1);
img_rain="./weather_image\rain\right\rain\rain_";
file = img_rain + num2str(r) +Image_end;
Rain_= imread(file);
end
if(Rain(id,3)==1)
alpha= randi([15, 25],1);
[distortion, D] = Global_rain(Name,D,I,Rain_,data,alpha*0.1,outputFolder);
end
if(Rain(id,3)==2)
alpha= randi([25, 40],1);
[distortion, D] = Global_rain(Name,D,I,Rain_,data,alpha*0.1,outputFolder);
end
if(Rain(id,3)==3)
alpha= randi([40, 55],1);
[distortion, D] = Global_rain(Name,D,I,Rain_,data,alpha*0.1,outputFolder);
end
end
end
end
else
D=imread(Dimg);
choice = randi([1,5],1);
if(choice==1)
angle = randi([1 360],1);
len = randi([1 15],1);
[distortion, D] = distortion_motion_blur(D,Image_name,len,angle,outputFolder);
end
% Check if global defocus distortion is selected and if so apply it
if(choice==2)
gaussian_size = 9;
gaussian_sigma = randi([1 20],1)*0.1;
[distortion, D] = distortion_defocus_blur(D,Image_name,gaussian_size,gaussian_sigma,outputFolder);
end
% Check if noise distortion is selected and if so apply it
if(choice==3)
means = 0;
variance = randi([1 20],1)*0.001;
[distortion, D] = distortion_noise(D,Image_name,means,variance,outputFolder);
end
% Check if compression distortion is selected and if so apply it
if(choice==4)
compression = randi([10 50],1);
[distortion, D] = distortion_compression(D,Image_name,compression,outputFolder);
end
% Check if contrast distortion is selected and if so apply it
if(choice==5)
cn = randi([1 10],1);
[distortion, D] = Global_contraste(Name, D,contrast(cn,:),outputFolder);
end
LK = [Image_name,' -- image number: '];
LK1 = [num2str(id),' -- distortion: '];
LK2 = [ distortion,' '];
displays= [ LK LK1 LK2];
disp(displays)
saving(id).Name=Name;
saving(id).Distortion=distortion;
end
% LK = [Image_name,' -- image number: '];
% LK1 = [num2str(id),' -- distortion: '];
% LK2 = [ distortion,' '];
% displays= [ LK LK1 LK2];
% disp(displays)
% saving(id).Name=Name;
% saving(id).Distortion=distortion;
end
time=toc
save('dist_data_val_test.mat','Global_distortion_choice');
% Analyze distortion distribution