Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
All padding functions to separate module
- Loading branch information
1 parent
ed19e71
commit 79f0dde
Showing
3 changed files
with
137 additions
and
130 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1 +1,2 @@ | ||
| from .pad import * | ||
| from .utils import * |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,136 @@ | ||
| import numpy as np | ||
|
|
||
| from .utils import degrees_to_pixels | ||
|
|
||
|
|
||
| def pad_img(img, padding, ppd, val): | ||
| """ | ||
| padding: degrees visual angle (top, bottom, left, right) | ||
| """ | ||
| padding_px = np.array(degrees_to_pixels(padding, ppd), dtype=np.int32) | ||
| padding_top, padding_bottom, padding_left, padding_right = padding_px | ||
| return np.pad( | ||
| img, | ||
| ( | ||
| (int(padding_top), int(padding_bottom)), | ||
| (int(padding_left), int(padding_right)), | ||
| ), | ||
| "constant", | ||
| constant_values=((val, val), (val, val)), | ||
| ) | ||
|
|
||
|
|
||
| def pad_img_to_shape(img, shape, val=0): | ||
| """ | ||
| shape: shape of the resulting image in pixels (height, width) | ||
| """ | ||
| height_px, width_px = shape | ||
| height_img_px, width_img_px = img.shape | ||
| if height_img_px > height_px or width_img_px > width_px: | ||
| raise ValueError("the image is bigger than the size after padding") | ||
|
|
||
| padding_vertical_top = int((height_px - height_img_px) // 2) | ||
| padding_vertical_bottom = int(height_px - height_img_px - padding_vertical_top) | ||
|
|
||
| padding_horizontal_left = int((width_px - width_img_px) // 2) | ||
| padding_horizontal_right = int(width_px - width_img_px - padding_horizontal_left) | ||
|
|
||
| return np.pad( | ||
| img, | ||
| ( | ||
| (padding_vertical_top, padding_vertical_bottom), | ||
| (padding_horizontal_left, padding_horizontal_right), | ||
| ), | ||
| "constant", | ||
| constant_values=val, | ||
| ) | ||
|
|
||
|
|
||
| def pad_array(arr, amount, pad_value=0): | ||
| """ | ||
| Pad array with an arbitrary value. So far, only works for 2D arrays. | ||
| Parameters | ||
| ---------- | ||
| arr : numpy ndarray | ||
| the array to be padded | ||
| amount : number or numpy ndarray | ||
| the amount of padding in each direction. Has to be of shape | ||
| len(arr.shape) X 2. the n-th row specifies the amount of padding | ||
| to be added to the n-th dimension of arr. The first value is the | ||
| amount of padding added before, the second value after the array. | ||
| If amount is a single number, it is used for padding in all | ||
| directions. | ||
| pad_value : number, optional | ||
| the value to be padded. Default is 0. | ||
| Returns | ||
| ------- | ||
| output : numpy ndarray | ||
| the padded array | ||
| """ | ||
| # if amount is a single number, use it for padding in all directions | ||
| if type(amount) is int or type(amount) is float: | ||
| amount = np.array(((amount, amount), (amount, amount))) | ||
| assert amount.amin() >= 0 | ||
| if len(arr.shape) != 2: | ||
| raise NotImplementedError("pad_array currently only works for 2D arrays") | ||
| if amount.sum() == 0: | ||
| return arr | ||
|
|
||
| output_shape = [x + y.sum() for x, y in zip(arr.shape, amount)] | ||
| output = np.ones(output_shape, dtype=arr.dtype) * pad_value | ||
| output[ | ||
| amount[0][0] : output_shape[0] - amount[0][1], | ||
| amount[1][0] : output_shape[1] - amount[1][1], | ||
| ] = arr | ||
| return output | ||
|
|
||
|
|
||
| def center_array(arr, shape, pad_value=0): | ||
| """Center an array on a larger one. Selects appropriate pad amounts in | ||
| every direction. | ||
| Parameters | ||
| ---------- | ||
| arr : numpy ndarray | ||
| the array to be padded | ||
| shape : tuple of two ints | ||
| the shape of the desired output array. Must be at least as large as | ||
| the input, and even for even input shapes, and odd for odd input | ||
| shapes. | ||
| pad_value : number, optional | ||
| the value to be padded. Default is 0. | ||
| Returns | ||
| ------- | ||
| output : numpy ndarray | ||
| the padded array | ||
| """ | ||
| if arr.shape == shape: | ||
| return arr | ||
| y_pad, x_pad = np.asarray(shape) - arr.shape | ||
| assert (y_pad % 2 == 0) and (x_pad % 2 == 0) | ||
| assert x_pad > 0 and y_pad > 0 | ||
| out = np.ones(shape, dtype=arr.dtype) * pad_value | ||
| out[y_pad / 2 : -y_pad / 2, x_pad / 2 : -x_pad / 2] = arr | ||
| return out | ||
|
|
||
|
|
||
| def resize_array(arr, factor): | ||
| """ | ||
| Return a copy of an array, resized by the given factor. Every value is | ||
| repeated factor[d] times along dimension d. | ||
| Parameters | ||
| ---------- | ||
| arr : 2D array | ||
| the array to be resized | ||
| factor : tupel of 2 ints | ||
| the resize factor in the y and x dimensions | ||
| Returns | ||
| ------- | ||
| An array of shape (arr.shape[0] * factor[0], arr.shape[1] * factor[1]) | ||
| """ | ||
| return np.repeat(np.repeat(arr, factor[0], axis=0), factor[1], axis=1) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters