CODEWITHSUNDEEP
imagematlabrgbpaste
fersarr
fersarr

Reputation: 3529

Matlab: paste RGB image into bigger image

I am trying to read an image and paste it into a bigger image in which I will, later, paste other images (same heights and widths). I have to say that I'm not experienced in Matlab so any suggestions are welcome.

Right now I'm creating a bigger matrix of zeroes and pasting the elements (RGB) of the image into it. But it is not being displayed as I would want, it shows a mostly-white image:

Bigger is the name of the bigger image

[im1 map1] = imread('/12937.png');
[height width rgbsize]=size(im1)
bigger=zeros(height+200,width+200,3);
figure('name','original');imshow(im1) %displays my image correctly
bigger(1:height,1:width,:)=im1(:,:,:); 
figure('name','after');imshow(bigger); %displays a mostly white image with dark right and bottom borders (the extra size)

Upvotes: 1

Views: 1727

Answers (1)

Bonlenfum
Bonlenfum

Reputation: 20165

Some of the image functions are sensitive to the data types. imread gives you a matrix of type uint8, whereas by default, zeros gives you matrices of type double. imshow (or image or imagesc) can operate with all double data, but they expect it $\in [0,1]$ rather than $\in [0,255]$.

Try this:

[im1 map1] = imread('/12937.png');
[height width rgbsize]=size(im1)

% note: initialise the data type as well as the size
bigger=zeros(height+200,width+200,3, 'uint8');

figure('name','original');imshow(im1) 
bigger(1:height,1:width,:)=im1(:,:,:); 
figure('name','after');imshow(bigger);

You can check what type im1 is with the command whos, or looking in the workspace part of the GUI. If it isn't uint8, then adjust the zeros command accordingly.

Upvotes: 1

Related Questions