Skip to content

Padding

Padding ¤

Bases: Subfunction_Base

A Padding Subfunction class which pads an images according to a desired shape.

Standard application is to square images to keep original aspect ratio. If another mode as "square" is selected, than a shape and NumPy pad mode is required!

Shape should be defined as tuple with x and y size:
Padding(mode="edge", shape=(224, 224))

Padding is done via NumPy pad function which uses can be called with different modes like "edge" or "minimum".
https://numpy.org/doc/stable/reference/generated/numpy.pad.html

Source code in aucmedi/data_processing/subfunctions/padding.py 
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
class Padding(Subfunction_Base):
    """ A Padding Subfunction class which pads an images according to a desired shape.

    Standard application is to square images to keep original aspect ratio.
    If another mode as "square" is selected, than a shape and NumPy pad mode is required!

    Shape should be defined as tuple with x and y size: <br>
    `Padding(mode="edge", shape=(224, 224))`

    Padding is done via NumPy pad function which uses can be called with different
    modes like "edge" or "minimum". <br>
    https://numpy.org/doc/stable/reference/generated/numpy.pad.html
    """
    #---------------------------------------------#
    #                Initialization               #
    #---------------------------------------------#
    def __init__(self, mode="square", shape=None):
        """ Initialization function for creating a Padding Subfunction which can be passed to a
            [DataGenerator][aucmedi.data_processing.data_generator.DataGenerator].

        Args:
            mode (str):                 Selected mode for image padding. If not `"square"`, then NumPy modes are used.
            shape (tuple of int):       Minimum image shape for non-`"square"` modes.
        """
        self.shape = shape
        self.mode = mode

    #---------------------------------------------#
    #                Transformation               #
    #---------------------------------------------#
    def transform(self, image):
        # Identify new shape
        if self.mode == "square":
            max_axis = max(image.shape[:-1])
            new_shape = [max_axis for x in range(0, len(image.shape[:-1]))]
        else:
            new_shape = [max(self.shape[i],image.shape[i]) \
                         for i in range(0, len(image.shape[:-1]))]
        # Compute padding width
        ## Code inspiration from: https://github.com/MIC-DKFZ/batchgenerators/blob/master/batchgenerators/augmentations/utils.py
        ## Leave a star for them if you are reading this. The MIC-DKFZ is doing some great work ;)
        difference = new_shape - np.asarray(image.shape[0:-1])
        pad_below = difference // 2
        pad_above = difference // 2 + difference % 2
        pad_list = list([list(i) for i in zip(pad_below, pad_above)]) + [[0, 0]]
        # Identify correct NumPy pad mode
        if self.mode == "square" : pad_mode = "edge"
        else : pad_mode = self.mode
        # Perform padding into desired shape
        image_padded = np.pad(image, pad_list, mode=pad_mode)
        # Return padded image
        return image_padded

__init__(mode='square', shape=None) ¤

Initialization function for creating a Padding Subfunction which can be passed to a DataGenerator.

Parameters:

Name Type Description Default
mode str

Selected mode for image padding. If not "square", then NumPy modes are used.

 'square'
shape tuple of int

Minimum image shape for non-"square" modes.

 None
Source code in aucmedi/data_processing/subfunctions/padding.py 
46
47
48
49
50
51
52
53
54
55
def __init__(self, mode="square", shape=None):
    """ Initialization function for creating a Padding Subfunction which can be passed to a
        [DataGenerator][aucmedi.data_processing.data_generator.DataGenerator].

    Args:
        mode (str):                 Selected mode for image padding. If not `"square"`, then NumPy modes are used.
        shape (tuple of int):       Minimum image shape for non-`"square"` modes.
    """
    self.shape = shape
    self.mode = mode