Showing an image with pylab.imshow()

Question

I'm relatively new to all this and I started the tutorial on image analysis here. When trying to perform the pylab.imshow(dna) step it returns the following error:

In [10]: pylab.imshow(dna)
---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-10-fc86cadb4e46> in <module>()
----> 1 pylab.imshow(dna)

 /usr/lib/pymodules/python2.7/matplotlib/pyplot.pyc in imshow(X, cmap, norm, aspect,    interpolation, alpha, vmin, vmax, origin, extent, shape, filternorm, filterrad, imlim, resample, url, hold, **kwargs)
   2375         ax.hold(hold)
   2376     try:
-> 2377         ret = ax.imshow(X, cmap, norm, aspect, interpolation, alpha, vmin, vmax, origin, extent, shape, filternorm, filterrad, imlim, resample, url, **kwargs)
   2378         draw_if_interactive()
   2379     finally:

/usr/lib/pymodules/python2.7/matplotlib/axes.pyc in imshow(self, X, cmap, norm, aspect, interpolation, alpha, vmin, vmax, origin, extent, shape, filternorm, filterrad, imlim, resample, url, **kwargs)
   6794                        filterrad=filterrad, resample=resample, **kwargs)
   6795 
-> 6796         im.set_data(X)
   6797         im.set_alpha(alpha)
   6798         self._set_artist_props(im)

/usr/lib/pymodules/python2.7/matplotlib/image.pyc in set_data(self, A)
    409         if (self._A.ndim not in (2, 3) or
    410             (self._A.ndim == 3 and self._A.shape[-1] not in (3, 4))):
--> 411             raise TypeError("Invalid dimensions for image data")
    412 
    413         self._imcache =None

TypeError: Invalid dimensions for image data

Fairly certain I have followed all the instructions in the tutorial to the letter but I can't work out was is going wrong.

Answer 1

it's just what the image is saved as in dna = mahotas.imread('dna.jpeg') type(dna) gives numpy.ndarray and dna.shape gives (1024, 1344, 1)

This is the problem, if you pass a 3D ndarray, it expects that you will have 3 or 4 planes (RGB or RGBA) (Read the code on line 410 in the last frame of the stack trace).

You just need to get rid of the extra dimension using

dna = dna.squeeze()

or

imshow(dna.squeeze())

To see what squeeze is doing, see the following example:

a = np.arange(25).reshape(5, 5, 1)
print a.shape # (5, 5, 1)
b = a.squeeze()
print b.shape # (5, 5)