stride_tricks: numpy.core._exceptions._ArrayMemoryError (How to accelerate the process of a custom template matching function)

Question

I have a custom template matching function that is based on Zero mean Normalized Cross Correlation: ZNCC formula

def ZNCC_mat(array1, array2):
    array1 = array1 / 255
    array2 = array2 / 255
    zmean_array1 = array1 - np.mean(array1)
    zmean_array1_norm = zmean_array1 / np.linalg.norm(zmean_array1)
    array2_view = np.lib.stride_tricks.sliding_window_view(
        array2, window_shape=array1.shape
    )
    zmean_array2 = array2_view - np.mean(array2_view, axis=(2, 3), keepdims=True)
    zmean_array2_view_norm = zmean_array2 / np.linalg.norm(
        zmean_array2, axis=(2, 3), keepdims=True
    )

    result = np.einsum("ij,klij->kl", zmean_array1_norm, zmean_array2_view_norm)

    return result

Quick explanation:

1. normalize the gray images to have values between 0-1
2. subtract the template by its mean and divide by the std
3. use np.lib.stride_tricks.sliding_window_view to divide the source image into sub images/matrices
4. subtract each region by its mean and divide by the std
5. use np.einsum to compute the cross correlation between the zero mean normalized template and each zero mean normalized region of the source

However after computing array2_view it is very heavy to do operations on the submatrices. I get this error:

Traceback (most recent call last):
  File "test/similarity_metrics_comparisons.py", line 17, in <module>
    result = ZNCC_mat(template_gray, source_gray)
  File "test/einsum_tests.py", line 145, in ZNCC_mat
    zmean_array2 = array2_view - np.mean(array2_view, axis=(2, 3), keepdims=True)
numpy.core._exceptions._ArrayMemoryError: Unable to allocate 35.4 GiB for an array with shape (478, 1150, 97, 89) and data type float64

Any tips on what could optimize the operations?

if you want to run the code, with drawing a rectangle around the estimated position:

example of images: source template

#do not forget to read your source and template images
source = cv2.imread()
template = cv2.imread()
source_gray = cv2.cvtColor(source, cv2.COLOR_BGR2GRAY)
template_gray = cv2.cvtColor(template, cv2.COLOR_BGR2GRAY)

result = ZNCC_mat(template_gray, source_gray)
print(result.max())
h, w = template_gray.shape
y, x = np.unravel_index(result.argmax(), result.shape)

top_left = (x, y)
bottom_right = (x + w, y + h)
cv2.rectangle(source, top_left, bottom_right, (0, 255, 0), 2)

plt.subplot(131),plt.imshow(source)
plt.subplot(132),plt.imshow(result)
plt.subplot(133),plt.imshow(template)
plt.show()