np.rot90() corrupts an opencv image

Question

When trying to rotate a landscape image to portrait, after applying the rotation, I cannot draw on the image.

img1 = cv2.imread('a.jpg')
cv2.circle(img1, tuple([10,10]),radius = 3, color = (255,0,0))

works fine.

Then I try:

img2 = np.rot90(img1,3)
cv2.circle(img2, tuple([10,10]),radius = 3, color = (255,0,0))

and I get the error:

TypeError: Layout of the output array img is incompatible with cv::Mat (step[ndims-1] != elemsize or step[1] != elemsize*nchannels)

Looking at type(img2) and img2.dtype it seems identical to img1. the dimensions also seem fine (the first two dimensions are flipped, the third stays "3")

BTW: this seems to work. (why?):

img2 = np.rot90(img1,3)
img3 = img2.copy()
cv2.circle(img3, tuple([10,10]),radius = 3, color = (255,0,0))

Answer 1

I had the same issue and never got to the bottom of it. The workaround I used was to do the image rotation/flipping with OpenCV instead, like:

# flip image vertically
img = cv2.flip(img, 0)

# flip image horizontally
img = cv2.flip(img, 1)

# transpose image
img = cv2.transpose(img)

Note that rotation is equivalent to doing a transpose and a flip.