process an image using a mask

i have an image that i segmented using the image segmenter app in matlab . i need to perform mathmatical operations on the segmented part only (which is all the values greater than zero) while maintaing the same image size .

the segment image function is giving me a mask and a segmented image how do i use this mask to process the pixels where the mask equals (1) and ignore the pixels where the mask equals(0) while maintaining the same image size . i tried the (M=feval (A(this is a function handle),O)) function followed by (M=M(BW)) but it seems to change the image size

clear
close all
clc
N               = 256 ;
I               = Phantom3D('modified shepp-logan',N) ;
I               = squeeze(I(:,128,:)) ;
[BW,O]          = segmentImage(I) ;

[Guillaume]

how do i use this mask to process the pixels where the mask equals

It's all going to depend on the processing function. Normally, the function has to be aware of the mask so you'd pass it as an input. If passing the mask is not an option of the function, then either you're out of luck and have to use another function, or you apply the function to the whole image and mask the result afterward.

[Shimaa Salam]

ok that is good to know . my function is is radon and iradon do thoes functions accept the use of a mask??

and if i use the mask to the result is it going to effect the time of the processing or the quality of image??

Answer 1

[Guillaume]

No radon and iradon do not know about masking. You will have to pass the whole image. I'm not very familiar with these functions but my understanding is that this would affect the result. However, I think that if you set the pixels outside the mask to zero, then the result won't be affected so

maskedimage = originalimage .* mask;
radon(maskedimage)

[Shimaa Salam]

thank you . but this solution altered the shape of the image

[Meirbek Mussatayev]

Yes I think that you can modify and use the following codes:

n = 200; %size of images

x = linspace(-1,1,n);

y = x;

dx = x(2) - x(1); %step in x - used for getting physical value of radon transform coordinate

[X, Y] = meshgrid(x, y);

R = sqrt(X .^ 2 + Y .^ 2); %matrix of radial distance from centre of grid

circular_mask = R < 1;

masked_image = raw_image .* circular_mask;