forked from prvn16/gpucoderdemo_lane_detection
-
Notifications
You must be signed in to change notification settings - Fork 0
/
RoboNouMiChiClass.m
1103 lines (911 loc) · 43 KB
/
RoboNouMiChiClass.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
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
classdef RoboNouMiChiClass < RoboNouMiChiClassOc
%UNTITLED Summary of this class goes here
% Detailed explanation goes here
properties
codepath;
end
methods
function obj = RoboNouMiChiClass(inputArg1,inputArg2)
%UNTITLED2 Construct an instance of this class
% Detailed explanation goes here
obj.Property1 = inputArg1 + inputArg2;
obj.videoPlayer = vision.DeployableVideoPlayer;
obj.videoReader = vision.VideoFileReader;
obj.videoWriter = vision.VideoFileWriter;
obj.peopleDetector = vision.PeopleDetector();
% J = imread('JapaneseMen.jpeg');
% J = imread('IMG_1106.jpg');
% J = imread('Japanese3.jpeg');
obj.frame = imread('visionteam1.jpg');
end
function genCode(obj,type,func,target,precision, genCodeOnlyValue, args)
cfg = coder.gpuConfig(type);
cfg.GenerateReport = true;
cfg.GenCodeOnly=genCodeOnlyValue;
cfg.TargetLang = 'C++';
cfg.Toolchain = ['NVIDIA CUDA (w/Microsoft Visual C++ 2015) | gmake (64-bit Windows)'];
codepath = "sumpurn/codegen/"+target+"/"+type+"/"+precision+"/"+func;
command = "codegen -d " + codepath + " " + " -args " + args + " -config cfg " + func;
disp(command);
obj.codepath = codepath;
eval(command);
end
function buildExamples(obj,examples)
for ii=1:length(examples)
if examples(ii)==1
xx = imread('CarPaintDefect.jpg');
% From imrect
pp = [18.9999999999999 92 1031 188];
pp1 = bbox2points(pp);
myroi = roipoly(xx,pp1(:,1),pp1(:,2));
xx1 = xx;
xx1(repmat(~myroi,[1,3]))=0;
imwrite(xx1,'CarPaintDefectIsolated.jpg');
obj.setFrame(xx1);
mc = maskClassOc(1,2);mc.channels={[55 100 -100 100 -100 100]}; [aa,bb] = obj.createMask(mc);imshowpair(obj.frame,bb,'montage');
imwrite(bb,'CarPaintDefectIdentified.jpg');
elseif examples(ii)==2
xx = imresize(imread('IMG_6582.JPG'),0.25);
imwrite(xx,'AircraftBodyChalkmarks.jpg');
obj.setFrame(xx);
mc = maskClassOc(1,2);mc.channels={[55 100 -20 0 -20 -1]}; [aa,bb] = obj.createMask(mc);imshowpair(obj.frame,bb,'montage');
imwrite(bb,'ChalkmarkingsDetected.jpg');
end
end
end
function I = detectPerson(obj)
%METHOD1 Summary of this method goes here
% Detailed explanation goes here
% outputArg = obj.Property1 + inputArg;
%filename = 'visionteam1.jpg';
[bboxes,scores] = step(obj.peopleDetector,obj.frame);
I = insertObjectAnnotation(obj.frame,'rectangle',bboxes,'Person ');
end
function I = humanFrame(obj)
% The body is represented by 15 points in three space, i.e. a vector of
% length 45. The data consists of F, five vectors describing the average
% female and M, five vectors describing the average male. Four linked
% segments, indexed by L, are the head, torso, arms, and legs.
% Initial view
f = figure;
f.CurrentAxes.set(gcf,'color','white', ...
'name','Eigenwalker', ...
'menu','none', ...
'numbertitle','off')
F = [];
M = [];
H = [];
load walkers % Load F and M, Females and Males -- and a surprise.
V = (F+M)/2; % Coefficient matrix.
X = reshape(V(:,1),15,3);
L = {[1 5],[5 12],[2 3 4 5 6 7 8],[9 10 11 12 13 14 15]}; % Links
p = {};
for k = 1:4
p{k} = line(X(L{k},1),X(L{k},2),X(L{k},3), ...
'marker','o', ...
'markersize',10, ...
'linestyle','-');
end
set(p{1},'tag','happy', ...
'userdata',zeros(1,3));
% f.CurrentAxes([-750 750 -750 750 0 1500])
f.CurrentAxes.XLim=[-750 750];
f.CurrentAxes.YLim = [ -750 750];
f.CurrentAxes.ZLim = [0 1500];
lift = get(gca,'position')+[0 .04 0 0];
% f.CurrentAxes.set('xtick',[], ...
% 'ytick',[], ...
% 'ztick',[], ...
% 'position',lift, ...
% 'clipping','off')
view(160,10)
myaxis = gca;
% Start walkin'...
period = 151.5751;
omega = 2*pi/period;
fps = 120; % 120 fps when s(1)=1
t = 0;
dt = 2*pi/omega/fps;
s = [1.0 1.0 1.0 1.0 1.0 1.0]';
steps = 0
% while get(stop,'value') == 0
while steps < 1000
steps = steps+1;
% s = cell2mat(get(sliders,'value'));
t = t + s(1)*dt;
c = [1 sin(omega*t) cos(omega*t) sin(2*omega*t) cos(2*omega*t)]';
c = [1; s(2:5).*c(2:5)];
V = (F+M)/2 + s(6)*(F-M)/2;
X = reshape(V*c,15,3);
H = get(p{1},'userdata');
e = ones(size(H,1),1);
XH = [X(e,:)+H; X(5,:)];
set(p{1},'xdata',XH(:,1), ...
'ydata',XH(:,2), ...
'zdata',XH(:,3))
for k = 2:4
set(p{k},'xdata',X(L{k},1), ...
'ydata',X(L{k},2), ...
'zdata',X(L{k},3));
end
p
pause(1/(s(1)*fps))
end;
end
function I = detectFaces(obj)
% Create a detector object.
faceDetector = vision.CascadeObjectDetector;
% Read input image.
I = imread('visionteam.jpg');
% Detect faces.
bboxes = step(faceDetector, I);
% Annotate detected faces.
IFaces = insertObjectAnnotation(I, 'rectangle', bboxes, 'Face');
figure, imshow(IFaces), title('Detected faces');
end
function I = detectUpperbody(obj)
% Create a detector object and set properties.
bodyDetector = vision.CascadeObjectDetector('UpperBody');
bodyDetector.MinSize = [60 60];
bodyDetector.MergeThreshold = 10;
% Read input image and detect upper body.
I = imread('visionteam.jpg');
bboxBody = step(bodyDetector, I);
% Annotate detected upper bodies.
IBody = insertObjectAnnotation(I, 'rectangle',bboxBody,'Upper Body');
figure, imshow(IBody), title('Detected upper bodies');
end
function I = recognizeFace(obj)
faceDB = imageSet('att_faces','recursive');
%% Show faces
figure;
montage(faceDB(3).ImageLocation);
title("Faces for a person recorded");
%% Query face vs All faces
queryFace = 1;
galleryImage = read(faceDB(queryFace),1);
figure;
faces = {};
for i=1:size(faceDB,2)
faces(i) = faceDB(i).ImageLocation(1);
end
galleryImage = insertObjectAnnotation(galleryImage, 'rectangle',[100 100, 100, 100],'Mathew');
subplot(1,2,1)
imshow(galleryImage);
subplot(1,2,2);
montage(faces);
I = faces(1);
end
function I = rangeFilter(obj,nbhood)
I = rangefilt(obj.frame,nbhood);
end
function I = detectHumanMovement(obj)
%create object of videoReader
vid = VideoReader('../panasonic/sample_b_1.mp4');
%find number of frames in video
% no_of_frames = vid.NumberOfFrames;
% fprintf('number of frames in video : %d \n', no_of_frames);
obj.frames{1} = read(vid,1);
obj.frames{2} = read(vid,50);
obj.frames{3} = read(vid,60);
frame1gray = rgb2gray(obj.frames{2}-obj.frames{1});
frame2gray = rgb2gray(obj.frames{3}-obj.frames{1});
I = imshowpair(frame1gray,frame2gray);
end
function tableout = importfile(obj,workbookFile,sheetName,startRow,endRow)
%IMPORTFILE スプレッドシ?[トからデ?[タをインポ?[ト
% DATA = IMPORTFILE(FILE) FILE という名前の Microsoft Excel スプレッドシ?[ト
% ファイルの?ﾅ?奄ﾌ???[クシ?[トからデ?[タを読み取り?Aテ?[ブルとして返します?B
%
% DATA = IMPORTFILE(FILE,SHEET) 指定された???[クシ?[トから読み取ります?B
%
% DATA = IMPORTFILE(FILE,SHEET,STARTROW,ENDROW)
% 指定した???[クシ?[トから指定した?s間隔で読み取ります?BSTARTROW と ENDROW
% を連続していない?s間隔で一致するサイズのスカラ?[またはベクトルの組として指定します?Bファイルの?ﾅ後まで読み取るには?A無限の ENDROW
% を指定します?B%
% 例:
% DefectImageList = importfile('DefectImageList.xlsx','並べ替え後',1,1000);
%
% XLSREAD も参?ﾆしてください?B
% MATLAB による自動?ｶ?ｬ 2017/11/05 01:34:00
%% 入力の取り扱い
% シ?[トが指定されていない?????A?ﾅ?奄ﾌシ?[トを読み?桙ﾝます
if nargin == 1 || isempty(sheetName)
sheetName = 1;
end
% ?sの始点と?I点が指定されていない?????A既定値を定義します
if nargin <= 3
startRow = 1;
endRow = 1000;
end
%% デ?[タのインポ?[ト
[~, ~, raw1] = xlsread(workbookFile, sheetName, sprintf('A%d:C%d',startRow(1),endRow(1)));
[~, ~, raw2] = xlsread(workbookFile, sheetName, sprintf('E%d:E%d',startRow(1),endRow(1)));
raw = [raw1,raw2];
for block=2:length(startRow)
[~, ~, tmpRawBlock1] = xlsread(workbookFile, sheetName, sprintf('A%d:C%d',startRow(block),endRow(block)));
[~, ~, tmpRawBlock2] = xlsread(workbookFile, sheetName, sprintf('E%d:E%d',startRow(block),endRow(block)));
raw = [raw;tmpRawBlock1,tmpRawBlock2]; %#ok<AGROW>
end
stringVectors = string(raw(:,[1,2,3,4]));
stringVectors(ismissing(stringVectors)) = '';
%% テ?[ブルの???ｬ
tableout = table;
%% インポ?[トした配列を列変?薄ｼに割り当てます
tableout.VarName1 = categorical(stringVectors(:,1));
tableout.VarName2 = stringVectors(:,2);
tableout.VarName3 = categorical(stringVectors(:,3));
tableout.VarName5 = categorical(stringVectors(:,4));
end
function [BW,maskedImage] = segmentImage(obj,X,pts,tol)
%segmentImage Segment image using auto-generated code from imageSegmenter App
% [BW,MASKEDIMAGE] = segmentImage(X) segments image X using auto-generated
% code from the imageSegmenter App. The final segmentation is returned in
% BW, and a masked image is returned in MASKEDIMAGE.
% Auto-generated by imageSegmenter app on 06-Nov-2017
%----------------------------------------------------
if length(obj.gaborX) == 0
obj.createGaborFeatures(X);
end
% テクスチャを使用した自動クラスタリング
sz = size(obj.gaborX);
im = reshape(obj.gaborX,sz(1)*sz(2),[]);
s = rng;
rng('default');
L = kmeans(im,2,'Replicates',2);
rng(s);
BW = L == 2;
BW = reshape(BW,[sz(1) sz(2)]);
% テクスチャを使用した塗りつぶし
for pp=1:length(pts)
pt=pts(pp);
row = pt(1);
column = pt(2);
tolerance = 2.500000e-01*tol;
normGaborX = sum((obj.gaborX - obj.gaborX(row,column,:)).^2,3);
normGaborX = mat2gray(normGaborX);
addedRegion = grayconnected(normGaborX, row, column, tolerance);
BW = BW | addedRegion;
end
% % テクスチャを使用した塗りつぶし
% row = 100;
% column = 100;
% tolerance = 2.500000e-01;
% normGaborX = sum((obj.gaborX - obj.gaborX(row,column,:)).^2,3);
% normGaborX = mat2gray(normGaborX);
% addedRegion = grayconnected(normGaborX, row, column, tolerance);
% BW = BW | addedRegion;
% Create masked image.
maskedImage = X;
% maskedImage(~BW) = 0;
maskedImage(repmat(~BW,[1 1 3])) = 0;
obj.maskedSegmentedImage = maskedImage;
obj.BW = BW;
end
function gaborFeatures = createGaborFeatures(obj,im)
if size(im,3) == 3
obj.frame = obj.prepLab(im);
else
obj.frame = im;
end
obj.frame = im2single(obj.frame);
imageSize = size(obj.frame);
numRows = imageSize(1);
numCols = imageSize(2);
wavelengthMin = 4/sqrt(2);
wavelengthMax = hypot(numRows,numCols);
n = floor(log2(wavelengthMax/wavelengthMin));
wavelength = 2.^(0:(n-2)) * wavelengthMin;
deltaTheta = 45;
orientation = 0:deltaTheta:(180-deltaTheta);
obj.gabor = gabor(wavelength,orientation);
gabormag = imgaborfilt(obj.frame(:,:,1),obj.gabor);
for i = 1:length(obj.gabor)
sigma = 0.5*obj.gabor(i).Wavelength;
K = 3;
gabormag(:,:,i) = imgaussfilt(gabormag(:,:,i),K*sigma);
end
% Increases liklihood that neighboring pixels/subregions are segmented together
X = 1:numCols;
Y = 1:numRows;
[X,Y] = meshgrid(X,Y);
featureSet = cat(3,gabormag,X);
featureSet = cat(3,featureSet,Y);
featureSet = reshape(featureSet,numRows*numCols,[]);
% Normalize feature set
featureSet = featureSet - mean(featureSet);
featureSet = featureSet ./ std(featureSet);
gaborFeatures = reshape(featureSet,[numRows,numCols,size(featureSet,2)]);
% Add color/intensity into feature set
gaborFeatures = cat(3,gaborFeatures,obj.frame);
obj.gaborX = gaborFeatures;
end
function showGabor(obj)
% figure;
% nn = size(obj.gabor)
% % subplot(nn(1),nn(2),1)
% for p = 1:length(obj.gabor)
% % subplot(nn(1),nn(2),p);
% obj.montage{p}=real(obj.gabor(p).SpatialKernel);
% lambda = obj.gabor(p).Wavelength;
% theta = obj.gabor(p).Orientation;
% title(sprintf('Re[h(x,y)], \\lambda = %d, \\theta = %d',lambda,theta));
% end
mm=5701;
pp = length(obj.gabor)/4
obj.montage = zeros(mm,mm,1,pp);
for p= 1:pp
rows = size(obj.gabor(p).SpatialKernel,1);
columns = rows;
obj.montage(1:rows,1:columns,1,p)= real(obj.gabor(p).SpatialKernel);
end
% cmap = colormap(gray);
% montage(obj.montage,cmap);
end
function [BW, maskedImage, Lout, cmap1] = makeSuperpixels(obj,im,N, labrange, isSuperTrue)
% im = imresize(image,0.1);
BW = im(:,:,1) > labrange(1) & im(:,:,1) < labrange(2) & im(:,:,2) > labrange(3) & im(:,:,2) < labrange(4) & im(:,:,3) > labrange(5) & im(:,:,3) < labrange(6);
% cmap1 = cmap;
BW = bwareaopen(BW, 100);
SE = strel('square',50);
BW = imdilate(BW,SE);
%
BW = bwmorph(BW,'majority', 200);
%
% % Determine the connected components:
% CC = bwconncomp(BW, 4);
% % Compute the area of each component:
% S = regionprops(CC, 'Area');
% % Remove small objects:
% L = labelmatrix(CC);
% BW = ismember(L, find([S.Area] >= 25)) & ismember(L, find([S.Area] < 20000));
maskedImage = im;
doMaskedImage = false;
if doMaskedImage
maskedImage(repmat(~BW,[1,1,3])) = 0;
maskedImage = lab2rgb(maskedImage);
end
Lout = [];
cmap1 = [];
if isSuperTrue
[L,N] = superpixels(im,N,'isInputLab',true);
% BW1 = boundarymask(L);
meanColor = zeros(size(im),'like',im);
% meanColor = zeros(N,3);
meanColor1 = zeros(N,3);
idx = label2idx(L);
numRows = size(im,1);
numCols = size(im,2);
for labelVal = 1:N
redIdx = idx{labelVal};
greenIdx = idx{labelVal}+numRows*numCols;
blueIdx = idx{labelVal}+2*numRows*numCols;
meanColor(redIdx) = mean(im(redIdx));
meanColor(greenIdx) = mean(im(greenIdx));
meanColor(blueIdx) = mean(im(blueIdx));
meanColor1(labelVal,1) = mean(im(redIdx));
meanColor1(labelVal,2) = mean(im(greenIdx));
meanColor1(labelVal,3) = mean(im(blueIdx));
end
numColors = 50;
[idx1,cmap] = kmeans(meanColor1,numColors,'replicates',2);
cmap1 = cmap;
cmap = lab2rgb(cmap);
Lout = zeros(size(im,1),size(im,2));
for i = 1:N
Lout(idx{i}) = idx1(i);
end
gg = cmap1(:,1) > labrange(1) & cmap1(:,1) < labrange(2) & cmap1(:,2) > labrange(3) & cmap1(:,2) < labrange(4) & cmap1(:,3) > labrange(5) & cmap1(:,3) < labrange(6);
% cmap1 = cmap;
cmap1(repmat(~gg,[1,1,3])) = 0;
% figure;
% imshow(Lout,lab2rgb(cmap1));
end
end
function [moviergb, moviep] = processMovie(obj,mov, frames, scale)
% if obj.profiling == true
% profile on -history;
% end
% scale = 0.25;
% obj.setFrame(imresize(mov(:,:,:,1),scale));
% mysize = size(obj.frame);
% mysize = [ mysize(1:2), length(frames)];
% moviep = zeros(mysize,'logical');
n = 3; m = 3;
for ff=frames
if isstruct(mov)
obj.setFrame(imresize(mov(ff).cdata,scale));
else
obj.setFrame(imresize(mov(:,:,:,ff),scale));
end
doSuperPixel = false;
mylab = rgb2lab(obj.frame);
if ff == 1
moviep = zeros([ size(obj.frame,1), size(obj.frame,2), length(frames)],'double');
moviergb = zeros([ size(obj.frame,1), size(obj.frame,2), 3, length(frames)],'uint8');
obj.background = mylab(:,:,1);
obj.frames{1}=obj.background;
end
BBCHANGE = abs(mylab(:,:,1) - obj.frames{1}) > 3;
BBCHANGE1 = abs(mylab(:,:,1) - obj.background) > 3;
BBCHANGE1 = bwareaopen(BBCHANGE1,1600);
BBCHANGE1 = imfill(BBCHANGE1,'holes');
[BBW, maskedImage, Lout, cmap] = obj.makeSuperpixels(mylab,100, [0 100 6 30 5 40],doSuperPixel);
doSkeleton = false;
if doSkeleton
skeleton = bwmorph(BBW,'skel',10);
end
% subplot(n,m,1)
% image(obj.frame);
% title('Original')
%
% subplot(n,m,2)
% imshow(BBW);
%
% title(ff+" : Mask")
if doSuperPixel
subplot(n,m,3)
image(Lout);
colormap(lab2rgb(cmap));
title('M2 : Superpixel')
end
doMaskedImage = false;
if doMaskedImage
subplot(n,m,4)
image(maskedImage);
title(ff +" : Masked Image")
end
if doSkeleton
subplot(n,m,5)
imshow(skeleton);
title(ff +" Skeleton ")
end
% F = getframe(gcf);
% [X, Map] = frame2im(F);
% moviep(:,:,:,ff) = imfuse(obj.frame,bb,'montage');
moviep(:,:,ff) = mylab(:,:,1);
zz = sum(BBCHANGE) < 5;
moviep(:,~zz,ff) = 0;
bg = obj.background;
bgNOT = obj.background;
bg(:,~zz) = 0;
bgNOT(:,zz) = 0;
alpha = 0.1;
beta = 0.0;
obj.background = alpha*bg + (1-alpha)*moviep(:,:,ff) + bgNOT;
% obj.background = alpha*bg*beta + (1-alpha)*moviep(:,:,ff) + bgNOT*(1-beta);
moviep(:,:,ff) = BBCHANGE1;
myrgb = obj.frame;
% imshow(myrgb)
myrgb(repmat(~BBCHANGE1,[1,1,3])) = 0;
% imshow(myrgb)
moviergb(:,:,:,ff) = myrgb;
% imshow(moviergb(:,:,:,ff));
obj.frames{1} = mylab(:,:,1);
end
% if obj.profiling == true
% obj.profile=profile('info');
% profile('off');
% end
end
function F = makeMovie(obj)
figure
Z = peaks;
surf(Z)
axis tight manual
ax = gca;
ax.NextPlot = 'replaceChildren';
loops = 40;
F(loops) = struct('cdata',[],'colormap',[]);
for j = 1:loops
X = sin(j*pi/10)*Z;
surf(X,Z)
drawnow
F(j) = getframe;
end
end
function out = segmentFMM(obj,X)
% Convert RGB image into L*a*b* color space.
% X = rgb2lab(vidFrame);
figure;
m = 2; n = 2;
for i=1:m
for j=1:n
subplot(m,n,(i-1)*m+j);
% Create empty mask.
mask = false(size(X,1),size(X,2));
% Flood fill
row = size(X,1)*0.9/m*i;
column = size(X,2)*0.9/n*j;
mask(row-10:row+10,column-10:column+10) = true;
tolerance = 1;
normX0 = sum((X - X(row,column,:)).^2,3);
normX = mat2gray(normX0);
weightImage = graydiffweight(normX, mask, 'GrayDifferenceCutoff', tolerance, 'RolloffFactor',.5);
% imshow(log(weightImage),[]);
[addedRegion, D] = imsegfmm(weightImage, column, row, 0.01);
sigma = 1.5;
W = gradientweight(X(:,:,1)/100.0, sigma, 'RolloffFactor', 3, 'WeightCutoff', 0.25);
[addedRegion1, D] = imsegfmm(W, mask, 0.01);
% imshow(D);
% BW1 = BW | addedRegion;
imshow(imfuse(addedRegion,addedRegion1,'montage'));
% imshow(BW1);
hold on;
plot(column+[0 size(X,2)], row*[1 1], 'r.', 'LineWidth', 1.5, 'MarkerSize',15);
end
end
end
function s = makeMovieFromFile(obj,filename,scale)
vidObj = VideoReader(filename);
vidHeight = vidObj.Height*scale;
vidWidth = vidObj.Width*scale;
s = struct('cdata',zeros(vidHeight,vidWidth,3,'uint8'),...
'colormap',[]);
k = 1;
while hasFrame(vidObj)
s(k).cdata = imresize(readFrame(vidObj),scale);
k = k+1;
end
end
% ?
% ????? ????? ????? ????? ???? ????
% ?? ??, ?? ??, ?? ??
% ????? ?????, ???? ????
% ?????? ?????? ??? ???
% ???? ???? ????? ???? ????
% ?? ??, ?????, ?????? ??????
% ??????????, ???? ????
% ??? ??? ????
% ? ?? ?????
function tracks = initializeTracks(obj)
% create an empty array of tracks
obj.tracks = struct(...
'id', {}, ...
'bbox', {}, ...
'kalmanFilter', {}, ...
'age', {}, ...
'totalVisibleCount', {}, ...
'consecutiveInvisibleCount', {});
end
function predictNewLocationsOfTracks(obj)
for i = 1:length(obj.tracks)
bbox = obj.tracks(i).bbox;
% Predict the current location of the track.
predictedCentroid = predict(obj.tracks(i).kalmanFilter);
% Shift the bounding box so that its center is at
% the predicted location.
predictedCentroid = int32(predictedCentroid) - bbox(3:4) / 2;
obj.tracks(i).bbox = [predictedCentroid, bbox(3:4)];
end
end
function [assignments, unassignedTracks, unassignedDetections] = ...
detectionToTrackAssignment(obj)
nTracks = length(tracks);
nDetections = size(centroids, 1);
% Compute the cost of assigning each detection to each track.
cost = zeros(nTracks, nDetections);
for i = 1:nTracks
cost(i, :) = distance(tracks(i).kalmanFilter, centroids);
end
% Solve the assignment problem.
costOfNonAssignment = 20;
[assignments, unassignedTracks, unassignedDetections] = ...
assignDetectionsToTracks(cost, costOfNonAssignment);
end
function updateAssignedTracks(obj)
numAssignedTracks = size(assignments, 1);
for i = 1:numAssignedTracks
trackIdx = assignments(i, 1);
detectionIdx = assignments(i, 2);
centroid = centroids(detectionIdx, :);
bbox = bboxes(detectionIdx, :);
% Correct the estimate of the object's location
% using the new detection.
correct(obj.tracks(trackIdx).kalmanFilter, centroid);
% Replace predicted bounding box with detected
% bounding box.
obj.tracks(trackIdx).bbox = bbox;
% Update track's age.
obj.tracks(trackIdx).age = tracks(trackIdx).age + 1;
% Update visibility.
obj.tracks(trackIdx).totalVisibleCount = ...
obj.tracks(trackIdx).totalVisibleCount + 1;
obj.tracks(trackIdx).consecutiveInvisibleCount = 0;
end
end
function updateUnassignedTracks(obj)
for i = 1:length(unassignedTracks)
ind = unassignedTracks(i);
obj.tracks(ind).age = obj.tracks(ind).age + 1;
obj.tracks(ind).consecutiveInvisibleCount = ...
obj.tracks(ind).consecutiveInvisibleCount + 1;
end
end
function deleteLostTracks(obj)
if isempty(obj.tracks)
return;
end
invisibleForTooLong = 20;
ageThreshold = 8;
% Compute the fraction of the track's age for which it was visible.
ages = [obj.tracks(:).age];
totalVisibleCounts = [obj.tracks(:).totalVisibleCount];
visibility = totalVisibleCounts ./ ages;
% Find the indices of 'lost' tracks.
lostInds = (ages < ageThreshold & visibility < 0.6) | ...
[obj.tracks(:).consecutiveInvisibleCount] >= invisibleForTooLong;
% Delete lost tracks.
obj.tracks = obj.tracks(~lostInds);
end
function displayTrackingResults(obj)
% Convert the frame and the mask to uint8 RGB.
frame = im2uint8(obj.frame);
mask = uint8(repmat(mask, [1, 1, 3])) .* 255;
minVisibleCount = 8;
if ~isempty(obj.tracks)
% Noisy detections tend to result in short-lived tracks.
% Only display tracks that have been visible for more than
% a minimum number of frames.
reliableTrackInds = ...
[obj.tracks(:).totalVisibleCount] > minVisibleCount;
reliableTracks = obj.tracks(reliableTrackInds);
% Display the objects. If an object has not been detected
% in this frame, display its predicted bounding box.
if ~isempty(reliableTracks)
% Get bounding boxes.
bboxes = cat(1, reliableTracks.bbox);
% Get ids.
ids = int32([reliableTracks(:).id]);
% Create labels for objects indicating the ones for
% which we display the predicted rather than the actual
% location.
labels = cellstr(int2str(ids'));
predictedTrackInds = ...
[reliableTracks(:).consecutiveInvisibleCount] > 0;
isPredicted = cell(size(labels));
isPredicted(predictedTrackInds) = {' predicted'};
labels = strcat(labels, isPredicted);
% Draw the objects on the frame.
frame = insertObjectAnnotation(frame, 'rectangle', ...
bboxes, labels);
% Draw the objects on the mask.
mask = insertObjectAnnotation(mask, 'rectangle', ...
bboxes, labels);
end
end
% Display the mask and the frame.
obj.maskPlayer.step(mask);
obj.videoPlayer.step(frame);
end
function createNewTracks(obj)
centroids = centroids(unassignedDetections, :);
bboxes = bboxes(unassignedDetections, :);
for i = 1:size(centroids, 1)
centroid = centroids(i,:);
bbox = bboxes(i, :);
% Create a Kalman filter object.
kalmanFilter = configureKalmanFilter('ConstantVelocity', ...
centroid, [200, 50], [100, 25], 100);
% Create a new track.
newTrack = struct(...
'id', nextId, ...
'bbox', bbox, ...
'kalmanFilter', kalmanFilter, ...
'age', 1, ...
'totalVisibleCount', 1, ...
'consecutiveInvisibleCount', 0);
% Add it to the array of tracks.
obj.tracks(end + 1) = newTrack;
% Increment the next id.
nextId = nextId + 1;
end
end
function trackFace(obj, filename, scale,channel )
% Create a cascade detector object.
if isempty(filename)
filename='visionface.avi';
scale=0.5;
end
faceDetector = vision.CascadeObjectDetector();
% Read a video frame and run the detector.
videoFileReader = vision.VideoFileReader(filename);
obj.frame = imresize(step(videoFileReader),scale);
bbox = step(faceDetector,obj.frame);
% Draw the returned bounding box around the detected face.
videoOut = insertObjectAnnotation(obj.frame,'rectangle',bbox,'Face');
figure, imshow(videoOut), title('Detected face');
% Get the skin tone information by extracting the Hue from the video frame
% converted to the HSV color space.
[hueChannel,~,~] = rgb2hsv(obj.frame);
xxlab = rgb2lab(obj.frame);
bw = imquantize(xxlab(:,:,2),[4 30]);
if isempty(channel)
channel = [0 100 4 50 5 50];
end
[BW maskedImage] = obj.createMask(channel);
% Display the L Channel data and draw the bounding box around the face.
% figure, imshow(hueChannel), title('Hue channel data');
% imshow(xxlab(:,:,1),[]), title('L channel data');
if isempty(bbox)
bbox = [100 100 50 50];
end
rectangle('Position',bbox(1,:),'LineWidth',2,'EdgeColor',[1 1 0])
% Detect the nose within the face region. The nose provides a more accurate
% measure of the skin tone because it does not contain any background
% pixels.
noseDetector = vision.CascadeObjectDetector('Nose', 'UseROI', true);
noseBBox = step(noseDetector, obj.frame, bbox(1,:));
if isempty(noseBBox)
noseBBox = [110 110 50 50];
end
% Create a tracker object.
tracker = vision.HistogramBasedTracker;
hblob = vision.BlobAnalysis('MinimumBlobArea', 100);
labch = 1;
% Initialize the tracker histogram using the Hue channel pixels from the
% nose.
% feature = hueChannel;
feature = rescale(xxlab(:,:,labch));
initializeObject(tracker, feature, noseBBox(1,:));
% Create a video player object for displaying video frames.
videoInfo = info(videoFileReader);
videoPlayer = vision.VideoPlayer('Position',[300 300 videoInfo.VideoSize+30]);
obj.background = xxlab(:,:,1);
obj.frames{1}=obj.background;
% Track the face over successive video frames until the video is finished.
while ~isDone(videoFileReader)
% Extract the next video frame
obj.frame = imresize(step(videoFileReader),scale);
[aa,bb] = obj.createMask(channel);
[~,CENTROID,BBOX] = step(hblob,aa);
% xxlab = rgb2lab(obj.frame);
%
%
% BBCHANGE = abs(xxlab(:,:,1) - obj.frames{1}) > 10;
% BBCHANGE1 = abs(xxlab(:,:,1) - obj.background) > 10;
% BBCHANGE1 = bwareaopen(BBCHANGE1,100);
% BBCHANGE1 = imfill(BBCHANGE1,'holes');
%
% [aa,bb] = obj.createMask;
%
%
%
% % RGB -> HSV
% [hueChannel,~,~] = rgb2hsv(obj.frame);
%
% % Track using the Hue channel data
% % feature = hueChannel;
% feature = rescale(xxlab(:,:,labch));
%
% bbox = step(tracker, feature);
%
% % bbox = step(tracker, hueChannel);
%
%
%
% bw2 = imquantize(xxlab(:,:,2),[4 40]);
% bw3 = imquantize(xxlab(:,:,3),[6 50]);
% bw = bw2==2 | bw3==2;
% bw = bwareaopen(bw,100);
% % bw = BBCHANGE1;
%
% videoFrame1 = obj.frame;
%
% zz = sum(BBCHANGE) < 3;
% % videoFrame(:,~zz) = 0;
% bg = obj.background;
% bgNOT = obj.background;
%
% bg(:,~zz) = 0;
% bgNOT(:,zz) = 0;
%
%
% alpha = 0.5;
% obj.background = alpha*bg + (1-alpha)*xxlab(:,:,1) + bgNOT;
%
%
% obj.frames{1} = xxlab(:,:,1);
%
%
%
%
% videoFrame1(repmat(BW,[1, 1, 3]))=0;
%
% % Insert a bounding box around the object being tracked
% videoOut = insertObjectAnnotation(videoFrame1,'rectangle',bbox,'Face');
% Display the annotated video frame using the video player object
videoOut = insertObjectAnnotation(bb,'rectangle',BBOX,'Skin');