Subtract across Numpy array

Question

I have a numpy array in the shape (244,244,3). Each holds a RGB image as numpy array. From each of the channels I want to subtract a value.

Of course I can easily do it with a for loop, but I assume that there is a faster way in numpy?

Any ideas?

Many thanks

Are you subtracting an array of values of the same size? In that case you can just do array1 - array2; as simple as that. If this is not the case, can you clarify the dimensions of the values you'd like to substract? — isosceleswheel
– isosceleswheel, Commented Apr 24, 2017 at 22:37
that shape suggests that it is rgb image. your reference to 'each', does that mean you have others with the same shape? — NaN
– NaN, Commented Apr 24, 2017 at 22:39

wim · Accepted Answer · 2017-04-24 22:39:28Z

5

You may simply subtract a 3-vector, numpy will broadcast it automatically.

Little demo with a 5x5 RGB image, subtracting 2 from red channel, 5 from green channel, 3 from blue channel:

>>> A = 10*np.ones((5,5,3), dtype=int)
>>> A -= [2, 5, 3]
>>> A[:,:,0]  # Red
array([[8, 8, 8, 8, 8],
       [8, 8, 8, 8, 8],
       [8, 8, 8, 8, 8],
       [8, 8, 8, 8, 8],
       [8, 8, 8, 8, 8]])
>>> A[:,:,1]  # Green
array([[5, 5, 5, 5, 5],
       [5, 5, 5, 5, 5],
       [5, 5, 5, 5, 5],
       [5, 5, 5, 5, 5],
       [5, 5, 5, 5, 5]])
>>> A[:,:,2]  # Blue
array([[7, 7, 7, 7, 7],
       [7, 7, 7, 7, 7],
       [7, 7, 7, 7, 7],
       [7, 7, 7, 7, 7],
       [7, 7, 7, 7, 7]])

answered Apr 24, 2017 at 22:39

wim

368k114 gold badges681 silver badges817 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Subtract across Numpy array

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related