image segmentation error (tuple index out of range)

[Anurag Phadke]

Hello,
I am trying to run "image segmentation" on the attached image. Code at:
https://gist.github.com/anuragphadke/7eb0e4398ef005632c6d

However, on running the above script, i get the following error:

Traceback (most recent call last):
  File "a.py", line 21, in <module>
    bw = closing(image > thresh, square(3))
  File "/Library/Python/2.7/site-packages/skimage/morphology/misc.py", line 37, in func_out
    return func(image, selem=selem, *args, **kwargs)
  File "/Library/Python/2.7/site-packages/skimage/morphology/grey.py", line 122, in func_out
    out_temp = func(image, selem, out=out_temp, *args, **kwargs)
  File "/Library/Python/2.7/site-packages/skimage/morphology/grey.py", line 349, in closing
    dilated = dilation(image, selem)
  File "/Library/Python/2.7/site-packages/skimage/morphology/misc.py", line 37, in func_out
    return func(image, selem=selem, *args, **kwargs)
  File "/Library/Python/2.7/site-packages/skimage/morphology/grey.py", line 251, in dilation
    nd.grey_dilation(image, footprint=selem, output=out)
  File "/usr/local/lib/python2.7/site-packages/scipy/ndimage/morphology.py", line 1295, in grey_dilation
    sz = footprint.shape[ii]
IndexError: tuple index out of range

Any idea what might be wrong here? I tried multiple images and multiple file formats (jpg / png etc.)

-anurag

[Emmanuelle Gouillart]

Hello Anurag,

my guess is that your image is an RGB image, so that its shape is
(n_rows, n_cols, 3). I could reproduce your problem using the RGB
astronaut image from scikit-image (data.astronaut()). ``threshold_otsu``
is not very well adapted to RGB images because it mixes together the
color channels by first raveling all pixels. [Note for dev team: this
should be better documented, and we should look for a thresholding
algorithm compatible with RGB images]

Therefore ``image > thresh`` is a 3-D array (or 2-D color, you can try
imshow on this array). You can try determining the threshold value on a
grayscale image if it's possible for you, using the as_grey keyword
argument in io.imread. Or you can determine a threshold for each color
channel.

Cheers,
Emma

[Stéfan van der Walt]

Hi, Emmanuelle

Please file a ticket so that we don't forget about this missing feature.

Thanks!
Stéfan

--
You received this message because you are subscribed to the Google Groups "scikit-image" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image...@googlegroups.com.
To post to this group, send an email to scikit...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/scikit-image/20160108074452.GA3070033%40phare.normalesup.org.
For more options, visit https://groups.google.com/d/optout.

[Emmanuelle Gouillart]

Done!
https://github.com/scikit-image/scikit-image/issues/1867

[Anurag]

thx emma + Stefan. Here's my temporary work-around:

Convert to Grayscale, save as JPG, run the code.

-anurag

Done!
https://github.com/scikit-image/scikit-image/issues/1867

--

You received this message because you are subscribed to a topic in the Google Groups "scikit-image" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/scikit-image/QP6sHolCVgs/unsubscribe.
To unsubscribe from this group and all its topics, send an email to scikit-image...@googlegroups.com.

To post to this group, send an email to scikit...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/scikit-image/20160108075923.GB3070033%40phare.normalesup.org.

For more options, visit https://groups.google.com/d/optout.

--

Twitter: @anuragphadke (https://twitter.com/#!/anuragphadke)