OpenCV function similar to matlab's “find”

Question

I am looking for a function in openCV to help me make masks of images.

for example in MATLAB:

B(A<1)=0;

or

B=zeros(size(A));

B(A==

Answer 1

OpenCV C++ supports the following syntax you might find convenient in creating masks:

Mat B= A > 1;//B(A<1)=0

or

Mat B = A==10;
B *= c;

which should be equivalent to:

B=zeros(size(A));
B(A==10)=c;

You can also use compare(). See the following OpenCV Documentation.

Answer 2

Some functions allow you to pass mask arguments to them. To create masks the way you describe, I think you are after Cmp or CmpS which are comparison operators, allowing you to create masks, by comparison to another array or scalar. For example:

im = cv.LoadImageM('tree.jpg', cv.CV_LOAD_IMAGE_GRAYSCALE)
mask_im = cv.CreateImage((im.width, im.height), cv.IPL_DEPTH_8U, 1)
#Here we create a mask by using `greater than 100` as our comparison
cv.CmpS(im, 100, mask_im, cv.CV_CMP_GT)
#We set all values in im to 255, apart from those masked, cv.Set can take a mask arg.
cv.Set(im, 255, mask=mask_im)
cv.ShowImage("masked", im)
cv.WaitKey(0)

Original im:

im after processing: