File size: 5,448 Bytes
a89d9fd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
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
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
# copyright (c) 2020 PaddlePaddle Authors. All Rights Reserve.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#    http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from __future__ import unicode_literals

from PIL import Image, ImageEnhance, ImageOps
import numpy as np
import random
import six


class RawRandAugment(object):
    def __init__(self,
                 num_layers=2,
                 magnitude=5,
                 fillcolor=(128, 128, 128),
                 **kwargs):
        self.num_layers = num_layers
        self.magnitude = magnitude
        self.max_level = 10

        abso_level = self.magnitude / self.max_level
        self.level_map = {
            "shearX": 0.3 * abso_level,
            "shearY": 0.3 * abso_level,
            "translateX": 150.0 / 331 * abso_level,
            "translateY": 150.0 / 331 * abso_level,
            "rotate": 30 * abso_level,
            "color": 0.9 * abso_level,
            "posterize": int(4.0 * abso_level),
            "solarize": 256.0 * abso_level,
            "contrast": 0.9 * abso_level,
            "sharpness": 0.9 * abso_level,
            "brightness": 0.9 * abso_level,
            "autocontrast": 0,
            "equalize": 0,
            "invert": 0
        }

        # from https://stackoverflow.com/questions/5252170/
        # specify-image-filling-color-when-rotating-in-python-with-pil-and-setting-expand
        def rotate_with_fill(img, magnitude):
            rot = img.convert("RGBA").rotate(magnitude)
            return Image.composite(rot,
                                   Image.new("RGBA", rot.size, (128, ) * 4),
                                   rot).convert(img.mode)

        rnd_ch_op = random.choice

        self.func = {
            "shearX": lambda img, magnitude: img.transform(
                img.size,
                Image.AFFINE,
                (1, magnitude * rnd_ch_op([-1, 1]), 0, 0, 1, 0),
                Image.BICUBIC,
                fillcolor=fillcolor),
            "shearY": lambda img, magnitude: img.transform(
                img.size,
                Image.AFFINE,
                (1, 0, 0, magnitude * rnd_ch_op([-1, 1]), 1, 0),
                Image.BICUBIC,
                fillcolor=fillcolor),
            "translateX": lambda img, magnitude: img.transform(
                img.size,
                Image.AFFINE,
                (1, 0, magnitude * img.size[0] * rnd_ch_op([-1, 1]), 0, 1, 0),
                fillcolor=fillcolor),
            "translateY": lambda img, magnitude: img.transform(
                img.size,
                Image.AFFINE,
                (1, 0, 0, 0, 1, magnitude * img.size[1] * rnd_ch_op([-1, 1])),
                fillcolor=fillcolor),
            "rotate": lambda img, magnitude: rotate_with_fill(img, magnitude),
            "color": lambda img, magnitude: ImageEnhance.Color(img).enhance(
                1 + magnitude * rnd_ch_op([-1, 1])),
            "posterize": lambda img, magnitude:
            ImageOps.posterize(img, magnitude),
            "solarize": lambda img, magnitude:
            ImageOps.solarize(img, magnitude),
            "contrast": lambda img, magnitude:
            ImageEnhance.Contrast(img).enhance(
                1 + magnitude * rnd_ch_op([-1, 1])),
            "sharpness": lambda img, magnitude:
            ImageEnhance.Sharpness(img).enhance(
                1 + magnitude * rnd_ch_op([-1, 1])),
            "brightness": lambda img, magnitude:
            ImageEnhance.Brightness(img).enhance(
                1 + magnitude * rnd_ch_op([-1, 1])),
            "autocontrast": lambda img, magnitude:
            ImageOps.autocontrast(img),
            "equalize": lambda img, magnitude: ImageOps.equalize(img),
            "invert": lambda img, magnitude: ImageOps.invert(img)
        }

    def __call__(self, img):
        avaiable_op_names = list(self.level_map.keys())
        for layer_num in range(self.num_layers):
            op_name = np.random.choice(avaiable_op_names)
            img = self.func[op_name](img, self.level_map[op_name])
        return img


class RandAugment(RawRandAugment):
    """ RandAugment wrapper to auto fit different img types """

    def __init__(self, prob=0.5, *args, **kwargs):
        self.prob = prob
        if six.PY2:
            super(RandAugment, self).__init__(*args, **kwargs)
        else:
            super().__init__(*args, **kwargs)

    def __call__(self, data):
        if np.random.rand() > self.prob:
            return data
        img = data['image']
        if not isinstance(img, Image.Image):
            img = np.ascontiguousarray(img)
            img = Image.fromarray(img)

        if six.PY2:
            img = super(RandAugment, self).__call__(img)
        else:
            img = super().__call__(img)

        if isinstance(img, Image.Image):
            img = np.asarray(img)
        data['image'] = img
        return data