performancepython-2.7numpyvectorizationcentroid

Quickly calculating centroid of binary numpy array

I have a numpy array of 0's and 1's (512 x 512). I'd like to calculate the centroid of the shape of 1's (they're all connected in one circular blob in the middle of the array).

for i in xrange(len(array[:,0])):
    for j in xrange(len(array[0,:])):
        if array[i,j] == 1:
            x_center += i
            y_center += j

count = (aorta == 1).sum()
x_center /= count
y_center /= count

Is there a way to speed up my calculation above? Could I use numpy.where() or something? Are there any python functions for doing this in parallel?

Solution

  • You can replace the two nested loops to get the coordinates of the center point, like so -

    x_center, y_center = np.argwhere(array==1).sum(0)/count