How to add frame to image ?

Hello everyone.

i would like to know how to combine the below images to be one ?

my interest is to add frames to image.

thank you.

[Rik]

What this is doing is overwriting specific pixels, so this should just be a matter of resizing, padding if needed, and using logical indexing to merge/overwrite. What have you tried so far? And can you post those two images as separate test images?

[IJumperQ NoNeed]

i hardcoded it using the below code,however, it's not good because i want the frame to be the foreground and the image in background

image = imread('G:\4th stage\Savanah\PreImageProcessing\ImageProcessing\frames\frame4.jpg');
image2 = imread('G:\4th stage\Savanah\PreImageProcessing\ImageProcessing\frames\Owl.jpg');
image = imresize(image,[320 480]);
image2 = imresize(image2,[220 380]);
[row2 , col2 , dim2] = size(image)
[row , col , dim] = size(image2)
xc =round(col/2);
yc =round(row/2);
for i=1 : row
    for j=1 : col
        image(i+40,j+55,1) = image2(i,j,1);
        image(i+40,j+55,2) = image2(i,j,2);
        image(i+40,j+55,3) = image2(i,j,3);
    end
end
imshow(image);

[Rik]

I will not help you if you insist on not attaching example images. You should pad the image and then use logical indexing to only overwrite the pixels that are white in your frame image.

[IJumperQ NoNeed]

sorry for misleading you , i'm new to matlab as well as image processing tech.

i added the example images as attachments

Answer 1

[Rik]

Something like the code below should work. There are many ways to resize your image, below you'll find two examples.

frame=imread('frame4.jpg');
im=imread('Owl.jpg');

%select all white pixels in the frame
mask=all(frame>250,3);

%optional: select only the middle biggest blob
tmp=bwlabel(mask);
mask=tmp==tmp(round(end/2),round(end/2));

if true
    %now we need to make the image the same size as the mask
    im=imresize(im,size(mask));
else
    %we could also resize it in a different way and pad it with zeros
    mask_corners=[find(any(mask,2),1,'first') find(any(mask,2),1,'last') ...
        find(any(mask,1),1,'first') find(any(mask,1),1,'last')];
    im_inner=imresize(im,1+mask_corners([2 4])-mask_corners([1 3]));
    im_pad=zeros([size(mask) 3],'like',frame);
    im=im_pad;
    im(mask_corners(1):mask_corners(2),...
        mask_corners(3):mask_corners(4),:)=im_inner;
    %or any other way that results in im being the same size as mask
end

%extend the mask to RGB
mask=repmat(mask,1,1,3);

%merge the images
out_im=frame;
out_im(mask)=im(mask);

imshow(out_im)

[IJumperQ NoNeed]

thank you , it was straight-forward for what i want