-
Notifications
You must be signed in to change notification settings - Fork 2
/
cas_feat_mat.m
176 lines (150 loc) · 5.05 KB
/
cas_feat_mat.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
tic
im=(imread('C:\Users\sushena\Downloads\gabor feature extraction\rail_gray.jpg'));
B=imread('C:\Users\sushena\Downloads\gabor feature extraction\rail_rgb.jpg');
im2=rgb2gray(B);
im_r=B(:,:,1);
im_g=B(:,:,2);
im_b=B(:,:,3);
load('C:\Users\sushena\Downloads\gabor feature extraction\rgb-superpix-labels.mat');
load('C:\Users\sushena\Downloads\gabor feature extraction\gray-superpix-labels.mat');
load('E:\course\DIP\project\gray_gabour.mat');
load('E:\course\DIP\project\rgb_gabour.mat');
load('E:\course\DIP\project\rgb_surf.mat');
load('E:\course\DIP\project\gray_surf.mat');
N=max(max(L2));
avg_int_gray=zeros(N,1);
avg_std_gray=zeros(N,1);
avg_int_neigh=zeros(N,1);
avg_std_neigh=zeros(N,1);
N2=max(max(L));
avg_int_rgb=zeros(N2,1);
avg_std_rgb=zeros(N2,1);
avg_int_neigh_rgb=zeros(N2,1);
avg_std_neigh_rgb=zeros(N2,1);
avg_int_r=zeros(N2,1);
avg_int_g=zeros(N2,1);
avg_int_b=zeros(N2,1);
L_pad=padarray(L,[5,5]);
L2_pad=padarray(L2,[5,5]);
for i=1:N
[r,c]=find(L2==i);
avg_int_gray(i)=uint8(mean2(im(r(1:end),c(1:end))));
r_mid=median(r);
c_mid=median(c);
window=L2_pad(r_mid:r_mid+10,c_mid:c_mid+10);
neigh=unique(window(window>0));
avg_int_neigh(i)=round(mean(avg_int_gray(neigh)));
avg_std_gray(i)=(std2(im(r(1:end),c(1:end))));
avg_std_neigh(i)=(mean(avg_std_gray(neigh)));
end
av_in_gr=zeros(N,2);
av_std_gr=zeros(N,2);
av_in_gr(1:N,1)=avg_int_gray;
av_in_gr(1:N,2)=avg_int_neigh;
av_std_gr(1:N,1)=avg_std_gray;
av_std_gr(1:N,2)=avg_std_neigh;
for i=1:N2
[r2,c2]=find(L==i);
avg_int_rgb(i)=uint8(mean2(im2(r2(1:end),c2(1:end))));
r_mid=median(r2);
c_mid=median(c2);
window=L_pad(r_mid:r_mid+10,c_mid:c_mid+10);
neigh=unique(window(window>0));
avg_int_neigh_rgb(i)=round(mean(avg_int_rgb(neigh)));
avg_int_r(i)=uint8(mean2(im_r(r2(1:end),c2(1:end))));
avg_int_g(i)=uint8(mean2(im_g(r2(1:end),c2(1:end))));
avg_int_b(i)=uint8(mean2(im_b(r2(1:end),c2(1:end))));
avg_std_rgb(i)=(std2(im2(r2(1:end),c2(1:end))));
avg_std_neigh_rgb(i)=(mean(avg_std_rgb(neigh)));
end
av_in_rgb=zeros(N2,2);
av_st_rgb=zeros(N2,2);
av_in_rgb(1:N2,1)=avg_int_rgb;
av_in_rgb(1:N2,2)=avg_int_neigh_rgb;
av_st_rgb(1:N2,1)=avg_std_rgb;
av_st_rgb(1:N2,2)=avg_std_neigh_rgb;
gray_pix=size(gabfeat_super_gray,1);
rgb_pix=size(gabfeat_super,1);
gray_surf_pix=size(gabfeat_super_gray,1);
rgb_surf_pix=size(gabfeat_super,1);
gab_dist=zeros(gray_pix,rgb_pix);
surf_dist=zeros(gray_surf_pix,rgb_surf_pix);
for i=1:gray_pix
for j=1:rgb_pix
gab_dist(i,j)=sum((gabfeat_super_gray(i)-gabfeat_super(j)).^2).^0.5;
end
end
mean_gab_dist=zeros(gray_pix,1);
ref_mat=cell(gray_pix,1);
for i=1:gray_pix
mean_gab_dist(i)=mean(gab_dist(i,:));
ref_mat{i}=find(gab_dist(i,:)<mean_gab_dist(i));
end
for i=1:gray_surf_pix
for j=1:rgb_surf_pix
surf_dist(i,j)=sum((average_surf_gray(i)-average_surf(j)).^2).^0.5;
end
end
mean_surf_dist=zeros(gray_pix,1);
corr_surf=cell(gray_surf_pix,1);
mat_ind=cell(gray_surf_pix,1);
ref_surf_mat=cell(gray_surf_pix,1);
for i=1:gray_surf_pix
corr_surf{i}=surf_dist(ref_mat{i}(1:end));
mean_surf_dist(i)=mean(corr_surf{i});
mat_ind{i}=find(corr_surf{i}<mean_surf_dist(i));
ref_surf_mat{i}=ref_mat{i}(mat_ind{i}(1:end));
end
int_dist=zeros(N,N2);
std_dist=zeros(N,N2);
for i=1:N
for j=1:N2
int_dist(i,j)=sum((av_in_gr(i)-av_in_rgb (j)).^2).^0.5;
std_dist(i,j)=sum((av_std_gr(i)-av_st_rgb (j)).^2).^0.5;
end
end
mean_int_dist=zeros(N,1);
corr_int=cell(N,1);
mat_ind_int=cell(N,1);
ref_int_mat=cell(N,1);
for i=1:N
corr_int{i}=int_dist(ref_surf_mat{i}(1:end));
mean_int_dist(i)=mean(corr_int{i});
mat_ind_int{i}=find(corr_int{i}<mean_int_dist(i));
ref_int_mat{i}=ref_surf_mat{i}(mat_ind_int{i}(1:end));
end
mean_std_dist=zeros(N,1);
corr_std=cell(N,1);
mat_ind_std=cell(N,1);
ref_std_mat=cell(N,1);
for i=1:N
corr_std{i}=std_dist(ref_int_mat{i}(1:end));
mean_std_dist(i)=mean(corr_std{i});
mat_ind_std{i}=find(corr_std{i}<mean_std_dist(i));
ref_std_mat{i}=ref_int_mat{i}(mat_ind_std{i}(1:end));
end
fin_match=zeros(N,1);
F=cell(N,1);
min_F=zeros(N,1);
for i=1:N
F{i}=((0.2.*(gab_dist(i,ref_std_mat{i}(1:end))))+(0.5.*(surf_dist(i,ref_std_mat{i}(1:end))))+(0.2.*(int_dist(i,ref_std_mat{i}(1:end))))+(0.1.*(std_dist(i,ref_std_mat{i}(1:end)))));
min_F(i)=min(F{i});
fin_match(i)=ref_std_mat{i}(F{i}==min_F(i));
end
fin_sp_int=cell(N,1);
for i=1:N
fin_sp_int{i}(1)=avg_int_r(fin_match(i));
fin_sp_int{i}(2)=avg_int_g(fin_match(i));
fin_sp_int{i}(3)=avg_int_b(fin_match(i));
end
for i=1:N
[row,col]=find(L2==i);
im(row,col,1)=fin_sp_int{i}(1);
im(row,col,2)=fin_sp_int{i}(2);
im(row,col,3)=fin_sp_int{i}(3);
end
imshow(uint8(im));
g_name=imread('E:\course\DIP\project\rail_rgb_gray.jpg');
fin_img=colorizeFun(g_name,im,2);
figure,imshow(fin_img);
toc;