Python: Efficient creation of binary image from line

Question

Given a line I would like to quickly create a binary image with two regions separated by the line. I am doing it as

rows, cols = pix_arr.shape
arr = [ func(i) for i in range(0,rows*cols)]
arr = np.array(arr,dtype = 'bool')
arr = arr.reshape(pix_arr.shape)

func(i) is

func = lambda i: (i / cols) - m*(i % cols)-c < 0

and pix_arr is some 2D numpy array. m and c are the slope and constant in the equation for a line.

I get output such as

Is there a faster way of doing this?

Answer 1

You can simply do something like:

row_idx = np.arange(rows)[:, None]
col_idx = np.arange(cols)

binary_img = (row_idx - m * col_idx) > c

This will get you close to what you want, though depending on how you define the axes for your line equation, you may need to flip the result upside down.