Optimize iteration throught numpy array

Question

I'm swapping values of a multidimensional numpy array in Python. But the code is too slow. Another thread says:

Typically, you avoid iterating through them directly. ... there's a good chance that it's easy to vectorize.

So, do you know a way to optimize the following code?

import PIL.Image
import numpy

pil_image = PIL.Image.open('Image.jpg').convert('RGB')
cv_image = numpy.array(pil_image)
# Convert RGB to BGR
for y in range(len(cv_image)):
    for x in range(len(cv_image[y])):
        (cv_image[y][x][0], cv_image[y][x][2]) = (cv_image[y][x][2],
            cv_image[y][x][0])

For an 509x359 image this last more than one second, which is way too much. It should perform it's task in no time.

on a side note, cv_image[y][x][0] is generally written cv_image[y, x, 0] in numpy/python. — Bi Rico
– Bi Rico, Commented Jul 3, 2012 at 19:12

eumiro · Accepted Answer · 2012-07-03 19:04:42Z

5

How about this single operation inverting the matrix along the last axis?

cv_image = cv_image[:,:,::-1]

answered Jul 3, 2012 at 19:04

eumiro

214k36 gold badges307 silver badges264 bronze badges

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

4 Comments

rynd Over a year ago

There's something strange with this. Both types (source and result) are numpy.ndarray. When iterating through them and comparing len(...) or the values (reversely), there can't be found a difference. But when trying to show the image with cv2.imshow it says: error: (-206) Unrecognized or unsupported array type (while without the conversion the image is displayed [with wrong colors]).

rynd Over a year ago

I use the line: new_cv_image = cv_image[:, :, ::-1]. If I change one value in cv_image: cv_image[0, 0, 2] = 128, it is also changed in new_cv_image, although id(...) returns a different value for each.

Bi Rico Over a year ago

You could try cv_image = cv_image[..., ::-1].copy(). imshow is probably expecting a continuous array in memory.

rynd Over a year ago

@Bago: I came to the same conclusion (source: mbk.ps.uci.edu/python.html , search "copy"). This works perfectly well! I measured 0.0 seconds (no time). I would say: Goal achieved.

Collectives™ on Stack Overflow

Optimize iteration throught numpy array

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related