-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathTrainArgumentComputer.m
173 lines (151 loc) · 7.7 KB
/
TrainArgumentComputer.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
function [TrainArgument]=TrainArgumentComputer(TrainSignature,TaskArgument)
distance_matrix=zeros(3,5,5);
dtw_result_matrix=cell(5,5);
distance_vector=zeros(3,10);
TrainArgument.sum_Mahalanobis_matrix=zeros(length(TaskArgument.select_feature));
TrainArgument.optionfun=zeros(1,1+TaskArgument.iterations);
dtw_argument.dis_type=1;
dtw_argument.dtw_type=TaskArgument.dtw_type_train;
dtw_argument.slope=TaskArgument.slope;
dtw_argument.select_feature=TaskArgument.select_feature;
dtw_argument.distance=TaskArgument.distance;
dtw_argument.DTW_feature=TaskArgument.DTW_feature;
if TaskArgument.Mah_type~=2
dtw_argument.mahalanobis_covariance_matrix_inv=diag(ones(1,length(TaskArgument.select_feature)));
for i=1:5
for j=i+1:5
dtw_result=DTWCompare(cell2mat(TrainSignature(i)),cell2mat(TrainSignature(j)),dtw_argument);
dtw_result_matrix(i,j)={dtw_result};
distance_matrix(1,i,j)=dtw_result.distance;
distance_matrix(2,i,j)=dtw_result.mean_distance_len;
distance_matrix(3,i,j)=dtw_result.mean_distance_sum_len;
if TaskArgument.distance==2
TrainArgument.sum_Mahalanobis_matrix=TrainArgument.sum_Mahalanobis_matrix+dtw_result.sum_mah_dis;
TrainArgument.optionfun(1)=TrainArgument.optionfun(1)+dtw_result.optionfun;
end
end
end
end
if TaskArgument.distance==2
for iterations_count=1:TaskArgument.iterations
if TaskArgument.Mah_type==3 || TaskArgument.Mah_type==4
TrainArgument.sum_Mahalanobis_matrix_dif=zeros(length(TaskArgument.select_feature));
for i=6:length(TrainSignature)
for j=1:5
dtw_result=DTWCompare(cell2mat(TrainSignature(i)),cell2mat(TrainSignature(j)),dtw_argument);
TrainArgument.sum_Mahalanobis_matrix_dif=TrainArgument.sum_Mahalanobis_matrix_dif+dtw_result.sum_mah_dis;
end
end
elseif TaskArgument.Mah_type==5
GetPointMahalanobisMatrix(TrainSignature,TaskArgument,dtw_result_matrix);
end
switch TaskArgument.Mah_type
case 1
temp_matrix=TrainArgument.sum_Mahalanobis_matrix;
case 3
temp_matrix=(TrainArgument.sum_Mahalanobis_matrix_dif\TrainArgument.sum_Mahalanobis_matrix)/TrainArgument.sum_Mahalanobis_matrix_dif;
case 4
temp_matrix=TrainArgument.sum_Mahalanobis_matrix-TrainArgument.sum_Mahalanobis_matrix_dif*TaskArgument.alpha;
end
if TaskArgument.Mah_type==1 || TaskArgument.Mah_type==3 || TaskArgument.Mah_type==4
TrainArgument.mahalanobis_covariance_matrix_inv=inv(temp_matrix/ (det(temp_matrix) ^ ((1/(length(TaskArgument.select_feature))))));
elseif TaskArgument.Mah_type==2
TrainArgument.mahalanobis_covariance_matrix_inv=TaskArgument.mahalanobis_covariance_matrix_inv;
end
dtw_argument.dis_type=2;
if TaskArgument.Mah_type_diag==1
temp_martix=TrainArgument.mahalanobis_covariance_matrix_inv.*diag(ones(1,length(TrainArgument.mahalanobis_covariance_matrix_inv)));
TrainArgument.mahalanobis_covariance_matrix_inv=inv(temp_martix)/(det(inv(temp_martix))^(1/length(temp_martix)));
end
dtw_argument.mahalanobis_covariance_matrix_inv=TrainArgument.mahalanobis_covariance_matrix_inv;
TrainArgument.sum_Mahalanobis_matrix=0;
for i=1:5
for j=i+1:5
dtw_result=DTWCompare(cell2mat(TrainSignature(i)),cell2mat(TrainSignature(j)),dtw_argument);
dtw_result_matrix(i,j)={dtw_result};
distance_matrix(1,i,j)=dtw_result.distance;
distance_matrix(2,i,j)=dtw_result.mean_distance_len;
distance_matrix(3,i,j)=dtw_result.mean_distance_sum_len;
if TaskArgument.distance==2
TrainArgument.sum_Mahalanobis_matrix=TrainArgument.sum_Mahalanobis_matrix+dtw_result.sum_mah_dis;
TrainArgument.optionfun(iterations_count+1)=TrainArgument.optionfun(iterations_count+1)+dtw_result.optionfun;
end
end
end
end
end
% PlotTrainStd(TrainSignature,dtw_result_matrix,TaskArgument);
for k=1:3
distance_vector_index=1;
for i=1:length(distance_matrix)
for j=1:i
distance_matrix(k,i,j)=distance_matrix(k,j,i);
dtw_result_matrix(i,j)=dtw_result_matrix(j,i);
end
for j=i+1:length(distance_matrix)
distance_vector(k,distance_vector_index)=distance_matrix(k,i,j);
distance_vector_index=distance_vector_index+1;
end
end
end
if TaskArgument.distance==2
TrainArgument.TrainSignatureWeigthPath_1=ComputeLocalWeight(TrainSignature,dtw_result_matrix,reshape(distance_matrix(1,:,:),[5 5]),TaskArgument,TrainArgument);
TrainArgument.TrainSignatureWeigthPath_2=ComputeLocalWeight(TrainSignature,dtw_result_matrix,reshape(distance_matrix(2,:,:),[5 5]),TaskArgument,TrainArgument);
TrainArgument.TrainSignatureWeigthPath_3=ComputeLocalWeight(TrainSignature,dtw_result_matrix,reshape(distance_matrix(3,:,:),[5 5]),TaskArgument,TrainArgument);
end
%generate the train siganture
TrainArgument.feature=zeros(3,6);
for k=1:3
temp_mat=reshape(distance_matrix(k,:,:),size(distance_matrix,2),size(distance_matrix,3));
%min mean
TrainArgument.feature(k,1)=mean(min(temp_mat(temp_mat~=0),[],2));
%max mean
TrainArgument.feature(k,2)=mean(max(temp_mat,[],2));
%template mean
TrainArgument.feature(k,3)=sum(temp_mat(sum(temp_mat,2)==min(sum(temp_mat,2)),:))/(length(TrainSignature)-1);
%mean
TrainArgument.feature(k,4)=mean(distance_vector(k,:));
%std
TrainArgument.feature(k,5)=std(distance_vector(k,:));
%template number
TrainArgument.feature(k,6)=find(sum(temp_mat,2)==min(sum(temp_mat,2)));
end
TrainArgument.segmentResult=cell(3,1);
TrainArgument.TrainSignaturePhaseWeigth=cell(3,1);
for k=1:3
TrainArgument.segmentResult(k)={GetKeyPoint(cell2mat(TrainSignature(TrainArgument.feature(k,6))),TaskArgument)};
TrainArgument.TrainSignaturePhaseWeigth(k)={ComputerPhaseWeight(TrainArgument.feature(k,6),dtw_result_matrix,TrainArgument.segmentResult(k),TaskArgument)};
end
TrainArgument.allsegmentResult=cell(3,5);
for k=1
for i=1:5
if i==TrainArgument.feature(k,6)
KeyPointVector_=cell2mat(TrainArgument.segmentResult);
KeyPointVector=KeyPointVector_.KeyPoint;
temp_segment=zeros(size(KeyPointVector));
for index_2=1:length(KeyPointVector)
temp_segment(index_2)=KeyPointVector(index_2);
end
else
if i<TrainArgument.feature(k,6)
dtw_result=cell2mat(dtw_result_matrix(i,TrainArgument.feature(k,6)));
KeyPointVector_=cell2mat(TrainArgument.segmentResult);
KeyPointVector=KeyPointVector_.KeyPoint;
temp_segment=zeros(size(KeyPointVector));
for index_2=1:length(KeyPointVector)
temp_segment(index_2)=dtw_result.path1(find(dtw_result.path2==KeyPointVector(index_2),1,'last'));
end
else
dtw_result=cell2mat(dtw_result_matrix(TrainArgument.feature(k,6),i));
KeyPointVector_=cell2mat(TrainArgument.segmentResult);
KeyPointVector=KeyPointVector_.KeyPoint;
temp_segment=zeros(size(KeyPointVector));
for index_2=1:length(KeyPointVector)
temp_segment(index_2)=dtw_result.path2(find(dtw_result.path1==KeyPointVector(index_2),1,'last'));
end
end
end
TrainArgument.allsegmentResult(k,i)={temp_segment};
end
end
end