「画像リストの CSVファイルの読み込み」に続く内容として，「画像の整形 (変換)」を示す。
「画像リストの CSVファイルの読み込み」は，reader を構築したところまで：
<![CDATA[$ source [パス]/venv/bin/activate
(venv) $ python
>>> import tensorflow as tf
>>> tf.enable_eager_execution()
>>> csv_name = '(パス)/all_image_paths.csv'
>>> csvfile = open(csv_name, 'r')
>>> import csv
>>> reader = csv.reader(csvfile, delimiter=',')]]>
以下，これの続き：
1. リサイズ
データの画像は，サイズがまちまちである。
モデルに合わせて，これらを定形にリサイズする。
5x3, 3x5, 4x3 の３つの画像を，3x3 にリサイズ：
<![CDATA[>>> def resize_image_raw(image_raw):
... image = tf.image.decode_image(image_raw, channels=3)
... image = tf.image.resize(image, [3, 3])
... return image
...
>>> def resize_image(path):
... image_raw = tf.io.read_file(path)
... return resize_image_raw(image_raw)
...
>>> for row in reader:
... path = row[0]
... image = resize_image(path)
... image
...
<tf.Tensor: shape=(3, 3, 3), dtype=float32, numpy=
array([[[ 79. , 72. , 64. ],
[138.33333 , 141.66667 , 135. ],
[110.666664, 115.666664, 109.666664]],
[[ 74. , 71. , 54. ],
[125.666664, 116.666664, 76. ],
[103.66667 , 82.66667 , 51.666668]],
[[ 66. , 59. , 31. ],
[126.666664, 98.666664, 52. ],
[108.66667 , 85.66667 , 43.666668]]], dtype=float32)>
<tf.Tensor: shape=(3, 3, 3), dtype=float32, numpy=
array([[[187. , 197. , 222. ],
[176. , 181. , 200. ],
[174. , 184. , 186. ]],
[[126.333336, 158.33333 , 197.33333 ],
[126.333336, 139.33333 , 170.66667 ],
[112.666664, 114.333336, 122. ]],
[[142.33333 , 169.33333 , 214.33333 ],
[152.33333 , 172.33333 , 199.33333 ],
[138.66667 , 136.66667 , 139.66667 ]]], dtype=float32)>
<tf.Tensor: shape=(3, 3, 3), dtype=float32, numpy=
array([[[120. , 93. , 46. ],
[126.666664, 97. , 54.666668],
[ 82. , 69.666664, 37.333332]],
[[106. , 89. , 59. ],
[127. , 116.333336, 82. ],
[ 88.33333 , 84.33333 , 62.666664]],
[[116. , 121. , 115. ],
[140.33333 , 140.66667 , 130.66667 ],
[102.33333 , 97.33333 , 87.33333 ]]], dtype=float32)>
>>> ]]>
2. 画素値の範囲 [0, 255] を，[m, n] に変換
ここでは，範囲 [0, 1] に変換する：
<![CDATA[>>> csvfile = open(csv_name, 'r')
>>> reader = csv.reader(csvfile, delimiter=',')
>>> for row in reader:
... path = row[0]
... image = resize_image(path)
... image /= 255.0
... image
...
<tf.Tensor: shape=(3, 3, 3), dtype=float32, numpy=
array([[[0.3098039 , 0.28235292, 0.25098038],
[0.5424836 , 0.5555555 , 0.52941173],
[0.4339869 , 0.45359474, 0.43006533]],
[[0.29019606, 0.27843136, 0.2117647 ],
[0.49281043, 0.4575163 , 0.2980392 ],
[0.40653595, 0.32418302, 0.20261437]],
[[0.2588235 , 0.23137254, 0.12156862],
[0.496732 , 0.38692808, 0.20392156],
[0.42614383, 0.33594772, 0.17124183]]], dtype=float32)>
<tf.Tensor: shape=(3, 3, 3), dtype=float32, numpy=
array([[[0.7333333 , 0.772549 , 0.8705882 ],
[0.69019604, 0.7098039 , 0.7843137 ],
[0.6823529 , 0.7215686 , 0.7294117 ]],
[[0.4954248 , 0.620915 , 0.77385616],
[0.4954248 , 0.5464052 , 0.669281 ],
[0.44183004, 0.448366 , 0.47843134]],
[[0.5581699 , 0.66405225, 0.8405228 ],
[0.5973856 , 0.67581695, 0.7816993 ],
[0.5437909 , 0.53594774, 0.54771245]]], dtype=float32)>
<tf.Tensor: shape=(3, 3, 3), dtype=float32, numpy=
array([[[0.4705882 , 0.36470586, 0.18039215],
[0.496732 , 0.38039213, 0.21437907],
[0.3215686 , 0.2732026 , 0.14640522]],
[[0.41568625, 0.3490196 , 0.23137254],
[0.4980392 , 0.45620912, 0.3215686 ],
[0.34640518, 0.3307189 , 0.2457516 ]],
[[0.45490193, 0.47450978, 0.45098037],
[0.55032676, 0.55163395, 0.51241827],
[0.40130717, 0.38169932, 0.34248364]]], dtype=float32)>
>>> ]]>
|