I have read through the documentation for copyTo() but am still confused on how this function would be applied to the following code. This anwer states that we can use the c
Sorry that I kind of misled you there. Although it works nicely in C++, I cannot find the binding in Python. You can, however, use numpy.copyto function.
Here is a small demo that shows that both method (bitwise_not
and copyto
) produce identical result.
import cv2
import numpy as np
# Create two images
im1 = np.zeros((100, 100, 3), np.uint8)
im1[:] = (255, 0, 0)
im2 = np.zeros((100, 100, 3), np.uint8)
im2[:] = (0, 255, 0)
# Generate a random mask
ran = np.random.randint(0, 2, (100, 100), np.uint8)
# List of images and masks
images = [im1, im2]
mask = [ran, 1-ran]
not_output = np.zeros((100, 100, 3), np.uint8)
copy_output = np.zeros((100, 100, 3), np.uint8)
for i in range(0, len(images)):
# Using the 'NOT' way
not_output = cv2.bitwise_not(images[i], not_output, mask=mask[i])
# Using the copyto way
np.copyto(copy_output, images[i], where=mask[i][:, :, None].astype(bool))
cv2.imwrite('not.png', 255 - not_output)
cv2.imwrite('copy.png', copy_output)
Note that an extra dimension was padded to the mask array so that it can be broadcasted.