-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathcorrelation_coefficient_match.m
92 lines (89 loc) · 3.31 KB
/
correlation_coefficient_match.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
%影像的纹理机构清晰，特征明显，信噪比较高
[filename_left,pathname_left]=uigetfile('.bmp','读取左像');%50为左像，49为右像
image_left=imread(strcat(pathname_left,filename_left));
[im_height_left,im_width_left,degree_left]=size(image_left);
[filename_right,pathname_right]=uigetfile('.bmp','读取右像');
image_right=imread(strcat(pathname_right,filename_right));
[im_height_right,im_width_right,degree_right]=size(image_right);
imshow(image_left);
%a=image_left(2,3)指的是两行三列 y=2,x=3
[x_orgin,y_orgin]=ginput;%x代表宽度，y代表高度
hold on;
plot(x_orgin,y_orgin,'r*');
%用5*5的目标窗口去搜索右像的大部分区域
correlation_p=0;
x_target=0;
y_target=0;
window_length=9;
N=window_length*window_length;
Sg=0;
Sgg=0;
for i=y_orgin-(window_length-1)/2:y_orgin+(window_length-1)/2
for j=x_orgin-(window_length-1)/2:x_orgin+(window_length-1)/2
pixel=double(image_left(round(i),round(j)));
Sg=Sg+pixel;
Sgg=Sgg+pixel^2;
end
end
for x_right=1+(window_length-1)/2:0.75*im_width_right
for y_right=1+(window_length-1)/2:im_height_right-(window_length-1)/2
%计算目标窗口与搜索窗口的相关系数
%先计算搜索窗口内的像素总和和平方和
Sgt=0;
Sgtgt=0;
Sggt=0;%两窗口对应元素乘积之和
for i=y_right-(window_length-1)/2:y_right+(window_length-1)/2
for j=x_right-(window_length-1)/2:x_right+(window_length-1)/2
pixel=double(image_right(round(i),round(j)));
Sgt=Sgt+pixel;
Sgtgt=Sgtgt+pixel^2;
pixel_ggt=double(pixel)*double(image_left(round(y_orgin-(y_right-i)),round(x_orgin-(x_right-j))));
Sggt=Sggt+pixel_ggt;
end
end
%计算相关系数
corre_p=(Sggt-Sg*Sgt/N)/(sqrt((Sgg-Sg^2/N)*(Sgtgt-Sgt^2/N)));
if(corre_p>correlation_p)
correlation_p=corre_p;
x_target=x_right;
y_target=y_right;
end
end
end
figure(2);
imshow(image_right);
hold on;
%计算x_target-1,y_target与x_target+1,y_target的相关系数，然后进行一个抛物线拟合
Sgt=0;
Sgtgt=0;
Sggt=0;%两窗口对应元素乘积之和
for i=y_target-(window_length-1)/2:y_target+(window_length-1)/2
for j=x_target-1-(window_length-1)/2:x_target-1+(window_length-1)/2
pixel=double(image_right(round(i),round(j)));
Sgt=Sgt+pixel;
Sgtgt=Sgtgt+pixel^2;
pixel_ggt=double(pixel)*double(image_left(round(y_orgin-(y_target-i)),round(x_orgin-(x_target-1-j))));
Sggt=Sggt+pixel_ggt;
end
end
%计算相关系数
corre_p_before=(Sggt-Sg*Sgt/N)/(sqrt((Sgg-Sg^2/N)*(Sgtgt-Sgt^2/N)));
Sgt=0;
Sgtgt=0;
Sggt=0;%两窗口对应元素乘积之和
for i=y_target-(window_length-1)/2:y_target+(window_length-1)/2
for j=x_target+1-(window_length-1)/2:x_target+1+(window_length-1)/2
pixel=double(image_right(round(i),round(j)));
Sgt=Sgt+pixel;
Sgtgt=Sgtgt+pixel^2;
pixel_ggt=double(pixel)*double(image_left(round(y_orgin-(y_target-i)),round(x_orgin-(x_target+1-j))));
Sggt=Sggt+pixel_ggt;
end
end
corre_p_next=(Sggt-Sg*Sgt/N)/(sqrt((Sgg-Sg^2/N)*(Sgtgt-Sgt^2/N)));
%抛物线拟合的x方向坐标拟合，实验中选取了一点，其横向三点的相关系数分别为
%0.7583,0.9351,0.7436 由此进行抛物线拟合
x_target=x_target-(corre_p_next-corre_p_before)/(2*(corre_p_next+corre_p_before-2*correlation_p));
%抛物线修正后的像点坐标更加准确
plot(x_target,y_target,'b*');
hold on;