15

I would like to see what my ImageDataGenerator yields to my network.

Edit:
removed the channel_shift_range, accidently left it in the code

Generator

genNorm = ImageDataGenerator(rotation_range=10, width_shift_range=0.1, 
       height_shift_range=0.1, zoom_range=0.1, horizontal_flip=True)

Get Batches

batches = genNorm.flow_from_directory(path+'train', target_size=(224,224),
      class_mode='categorical', batch_size=64)

x_batch, y_batch = next(batches)

Plot Images

for i in range (0,32):
    image = x_batch[i]
    plt.imshow(image.transpose(2,1,0))
    plt.show()

Result

Generator Output

Is this normal or am I doing something wrong here?

Kay
  • 173
  • 1
  • 1
  • 9

3 Answers3

2

The strange colors result from your channel shift. Do you really need that to augment your samples? Is a value of 10 (=very high) really what you want?

In addition to that: Another and likely more efficient way of checking what your generator yields is to set a directory with save_to_dir (parameter of flow/flow from directory function). In that you´ll find all the images that have been transformed and been delivered to your fit/flow function.

Edit:

You still somehow seem to invert your images during processing or while displaying them. I assume the original images look more like this:

enter image description here

Save the results of your ImageDataGenerator to a directory and compare these with the results that you see with plt.show.

petezurich
  • 9,280
  • 9
  • 43
  • 57
  • Yeah, i tried a lot and left it accidently in the code... still the same issue. – Kay Aug 21 '17 at 13:36
  • If you save to a directory with save_to_dir and compare that to your plt.imshow screenshots – do the images look the same? You still seem to have inverted the images (see edit). – petezurich Aug 21 '17 at 20:04
  • You may have normalized the images by diving by 255 prior to visualizing them – Parseltongue Dec 08 '18 at 23:53
2

Try this; change the generator as follow:

import numpy as np

def my_preprocessing_func(img):
    image = np.array(img)
    return image / 255

genNorm = ImageDataGenerator(rotation_range=10, width_shift_range=0.1, 
       height_shift_range=0.1, zoom_range=0.1, horizontal_flip=True, 
       preprocessing_function=my_preprocessing_func)

That worked for me,

Bruno

user3452266
  • 21
  • 3
0

keras uses the image operation with Pillow backend, which comes with 'RGB' format default. So you don't need to reverse the channel axis in your plt.imshow().

Only in cv2.imread() (which is 'BGR' default), you may need plt.imshow(img[:, :, ::-1]) to display the right image.

BR

Nick
  • 138,499
  • 22
  • 57
  • 95
Sherk
  • 1
  • 1