images_alignment.utils module

Utilities functions or Classes

class images_alignment.utils.Terminal

Bases: object

Class to ‘write’ into the console

write(message)

Write message into the console

images_alignment.utils.fnames_multiframes(fname)

Return/create fnames in the TMP_DIR if the .tif has multiple frames

images_alignment.utils.gray_conversion(img)

Convert RGBA or RGB image to gray image

images_alignment.utils.rescaling_factor(imgs, max_size)

Return a ‘global’ rescaling factor satisfying ‘max_size’

images_alignment.utils.rescaling_factors(imgs, max_size)

Return the rescaling factors satisfying max_size for both item of imgs

images_alignment.utils.imgs_rescaling(imgs, max_size)

Rescale images according to ‘max_size’

images_alignment.utils.imgs_conversion(imgs)

Uniformize the number of dimension/channels (for images composition)

images_alignment.utils.image_normalization(img)

Normalize image in range [0., 1.]

images_alignment.utils.absolute_threshold(img, relative_threshold)

Return the absolute threshold to use when binarizing a ‘img’

images_alignment.utils.resizing(img1, img2)

Resize the images to have similar shape (requested for pyStackReg)

images_alignment.utils.cropping(img, area, verbosity=True)

Return cropped image according to the given area

images_alignment.utils.padding(img1, img2)

Add image padding

class images_alignment.utils.TranslationTransform

Bases: ProjectiveTransform

estimate(src, dst)

Estimate the transformation from a set of corresponding points

inverse()

Return the inverse translation.

images_alignment.utils.sift(img1, img2, model_class=None)

SIFT feature detection and descriptor extraction

Parameters:

  • img1, img2 (numpy.ndarray((m, n)), numpy.ndarray((p, q))) – The input images

  • model_class (Objet) – ‘model_class’ used by RANSAC. If None, consider AffineTransform from skimage.transform.

Returns:

  • tmat (numpy.ndarrays((3, 3))) – The related transformation matrix

  • points (list of 2 numpy.ndarray((n, 2)) – Keypoints coordinates as (row, col) related to the 2 input images.

images_alignment.utils.concatenate_images(img1, img2, alignment='horizontal')

concatenate img1 and img2

Parameters:

  • img1 ((m, n [, 3]) array) – First grayscale or color image.

  • img2 ((p, q [, 3]) array) – Second grayscale or color image.

  • alignment ({‘horizontal’, ‘vertical’}, optional) – Whether to show images side by side, 'horizontal', or one above the other, 'vertical'.

images_alignment.utils.rescaling(img, rfac=0.25)

Return image with the rescaling_factor applied in the 2 dimensions