1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
|
"""Common utility functions."""
from __future__ import absolute_import
import numpy as np
import scipy.ndimage
from ssim.compat import ImageOps
def convolve_gaussian_2d(image, gaussian_kernel_1d):
"""Convolve 2d gaussian."""
result = scipy.ndimage.correlate1d(
image, gaussian_kernel_1d, axis=0)
return scipy.ndimage.correlate1d(
result, gaussian_kernel_1d, axis=1)
def get_gaussian_kernel(gaussian_kernel_width=11, gaussian_kernel_sigma=1.5):
"""Generate a gaussian kernel."""
# 1D Gaussian kernel definition
gaussian_kernel_1d = np.arange(0, gaussian_kernel_width, 1.)
gaussian_kernel_1d -= gaussian_kernel_width / 2
gaussian_kernel_1d = np.exp(-0.5 * gaussian_kernel_1d**2 /
gaussian_kernel_sigma**2)
return gaussian_kernel_1d / np.sum(gaussian_kernel_1d)
def to_grayscale(img):
"""Convert PIL image to numpy grayscale array and numpy alpha array.
Args:
img (PIL.Image): PIL Image object.
Returns:
(gray, alpha): both numpy arrays.
"""
gray = np.asarray(ImageOps.grayscale(img)).astype(float)
imbands = img.getbands()
alpha = None
if 'A' in imbands:
alpha = np.asarray(img.split()[-1]).astype(float)
return gray, alpha
|
