Stitching image in 2 direction

Hello everyone.

Anyone know how to sitching 2 images in 2 direction.

2 pictures below can not stitching in 1 direction (the postion of the nose and eyes are not in the same column and rows)

Thanks in advance.

Answer 1

[Image Analyst]

If you don't want to butt the images against each other but there is some overlap that needs to be determined, then you will need to look up "panorama" : https://www.mathworks.com/matlabcentral/answers/?term=tag%3A%22panorama%22

See this Mathworks Link: https://www.mathworks.com/help/vision/examples/feature-based-panoramic-image-stitching.html

and this one: https://www.mathworks.com/discovery/image-registration.html

and this one: https://www.crisluengo.net/archives/76

[Hieu Tran Doan Trung]

Many usefull information, thanks so much for sharing this. I tried some of them.

At the moment i haven't solve the problem and i will stick with it in long period of time.

Anyway, i posted another question about the seams and stitching line in stitching image you may see it in the stitching pictures. Could you accept the questions please.

Many thanks to you. Good luck and have a nice day.

Answer 2

[Image Analyst]

To stitch images:

wideImage = [image1, image2]; % Must have same number of rows.
tallImage = [image1; image2]; % Must have same number of columns.

For a more general solution, see https://www.mathworks.com/matlabcentral/fileexchange/25797-stitch

[Hieu Tran Doan Trung]

Hello, thanks for quickly reply me.

Anyway, it is not simple like those solution above which can easily solve by montage or someothers thing.

2 pictures stitching together look like the pictures under. However, the pictures below is 1 direction stitching, not 2 direction stitching. I hope someone may give me a suggestion to solve this.

Thank you so much.

[Hieu Tran Doan Trung]

This is the code file for 1 direction stitching

clc;

clear all;

close all;

I1 = imread('Lena_p4.png');

I2 = imread('Lena_p5.png');

F = im2double(I1);

S = im2double(I2);

[rows cols] = size(F(:,:,1));

Tmp = [];

temp = 0;

for j = 1:cols% to prevent j to go beyond boundaries.

for i = 1:rows

temp = temp + (F(i,j,1) - S(i,1,1))^2;

end

Tmp = [Tmp temp]; % Tmp keeps growing, forming a matrix of 1*cols

temp = 0;

end

%

[Min_value, Index] = min(Tmp);

n_cols = Index + cols - 1;% New column of output image.

Opimg = [];

for i = 1:rows

for j = 1:Index

for y = 1:3

Opimg(i,j,y) = F(i,j,y);% First image is pasted till Index.

end

end

for k = Index+1:n_cols-1

for y = 1:3

Opimg(i,k,y) = S(i,k-Index+1,y);%Second image is pasted after Index.

end

end

end

[r_Opimg c_Opimg] = size(Opimg(:,:,1));

subplot(1,3,1);

imshow(F);axis ([1 c_Opimg 1 c_Opimg])

title('First Image');

subplot(1,3,2);

imshow(S);axis ([1 c_Opimg 1 c_Opimg])

title('Second Image');

subplot(1,3,3);

imshow(Opimg);axis ([1 c_Opimg 1 c_Opimg])