Mailman 3 December 2015 - scikit-image

New to Contributing
by Kshitij Saraogi Dec. 3, 2015

Dec. 3, 2015

Hello, I am Kshitij Saraogi, a second year under-graduate student at IIT Kharagpur. I find scikit-image really intriguing and I would like to contribute to it. I went through the Issues section on GitHub but I am having a hard time finding an issue to get started Also, while going through the list of "Requested Features", I read about Image Colorisation. I would like to work on this. So, I would really appreciate if someone can guide me. Thanks Kshitij Saraogi

3 4

Re: label_boundaries problem with ndim = label_img.ndim
by Juan Nunez-Iglesias Dec. 2, 2015

Dec. 2, 2015

You don't need to find the boundaries manually before calling mark_boundaries. From the docstring of mark_boundaries: Parameters ---------- image : (M, N[, 3]) array Grayscale or RGB image. label_img : (M, N) array of int Label array where regions are marked by different integer values. Currently, you are passing a boolean array to a function that expects an integer array, which is why it's failing. In general, it's a good idea to try typing "help(function_name)" when things are failing to make sure that you are passing in the right information to the function you're using! (You can then press "q" to exit help.) I hope that, erm, helps! =) Juan. On Wed, Dec 2, 2015 at 6:46 PM, Eyal Saiet <ejsaiet(a)alaska.edu> wrote: > thanks Juan, I have tried that too: > > 1. label_image=measure.label(g_median_bin, background=0) > 2. label_image +=1 > 3. plt.figure() > 4. plt.imshow(label_image,cmap=plt.cm.gray,interpolation='nearest', > origin='lower') > 5. plt.title('label_image') > 6. plt.show() > 7. props=measure.regionprops(label_image,intensity_image=g_median_bin) > 8. label_bound=segmentation.find_boundaries(label_image,connectivity=1, > mode='thick',background=0) > 9. boundaries=segmentation.mark_boundaries(g_median,label_bound) > > Traceback (most recent call last): > File attributes_thrash.py", line 127, in <module> > > boundaries=segmentation.mark_boundaries(g_median,label_bound)#,color=(1,0,0), > outline_color=None, mode='outer', background_label=0) > File > "/usr/local/lib/python3.4/dist-packages/skimage/segmentation/boundaries.py", > line 213, in mark_boundaries > background=background_label) > File > "/usr/local/lib/python3.4/dist-packages/skimage/segmentation/boundaries.py", > line 156, in find_boundaries > max_label = np.iinfo(label_img.dtype).max > File "/usr/local/lib/python3.4/dist-packages/numpy/core/getlimits.py", > line 254, in __init__ > raise ValueError("Invalid integer data type.") > ValueError: Invalid integer data type. > > Process finished with exit code 1 > > > On Tue, Dec 1, 2015 at 10:38 PM, Juan Nunez-Iglesias <jni.soma(a)gmail.com> > wrote: > >> Hi Eyal, >> >> You are passing a list of regionprops (props, in your code) to >> "find_boundaries", whereas it expects an array: a single image of labeled >> objects (label_img, in your code). >> >> Juan. >> >> On Wed, Dec 2, 2015 at 5:33 PM, Eyal Saiet <ejsaiet(a)alaska.edu> wrote: >> >>> Stefan, >>> I do not understand what is not true about the label_image object. >>> bellow is the code, in addition I added a plot of the label_image object. >>> >>> >>> >>> 1. label_image=measure.label(g_median_bin, background=0) >>> 2. label_image +=1 >>> 3. plt.figure() >>> 4. plt.imshow(label_image,cmap=plt.cm.gray,interpolation='nearest', >>> origin='lower') >>> 5. plt.title('label_image') >>> 6. plt.show() >>> 7. props = measure.regionprops(label_image,intensity_image >>> =g_median_bin) >>> 8. label_bound=segmentation.find_boundaries(props,connectivity=1,mode >>> ='thick',background=0) >>> 9. boundaries=segmentation.mark_boundaries(g_median,label_bound) >>> >>> >>> >>> >>> [image: Inline image 1] >>> >>> On Tue, Dec 1, 2015 at 8:09 PM, Stefan van der Walt < >>> stefanv(a)berkeley.edu> wrote: >>> >>>> On 2015-12-01 21:00:16, Arctic_python <ejsaiet(a)alaska.edu> wrote: >>>> > ndim = label_img.ndim >>>> > AttributeError: 'list' object has no attribute 'ndim' >>>> > >>>> > 1. label_image=measure.label(g_median_bin, background=0) # >>>> g_median_bin >>>> > is a binary image >>>> >>>> That does not seem to be true, given the error message. >>>> >>>> Stéfan >>>> >>>> -- >>>> 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/PwmizxFahYE/unsubscribe. >>>> To unsubscribe from this group and all its topics, send an email to >>>> scikit-image+unsubscribe(a)googlegroups.com. >>>> To post to this group, send an email to scikit-image(a)googlegroups.com. >>>> To view this discussion on the web, visit >>>> https://groups.google.com/d/msgid/scikit-image/87zixt8mie.fsf%40berkeley.edu >>>> . >>>> For more options, visit https://groups.google.com/d/optout. >>>> >>> >>> >>> >>> -- >>> Eyal Saiet >>> >>> Project manager >>> Remote sensing and in-situ measurements >>> >>> Geophysical Institute >>> University of Alaska Fairbanks >>> Fairbanks, AK 99775 >>> (907) 750 6555 (cell) >>> >>> ejsaiet(a)alaska.edu >>> >>> -- >>> 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+unsubscribe(a)googlegroups.com. >>> To post to this group, send email to scikit-image(a)googlegroups.com. >>> To view this discussion on the web, visit >>> https://groups.google.com/d/msgid/scikit-image/CA%2BysReAKgzF%3DJXEjz7hti8s… >>> <https://groups.google.com/d/msgid/scikit-image/CA%2BysReAKgzF%3DJXEjz7hti8s…> >>> . >>> >>> For more options, visit https://groups.google.com/d/optout. >>> >> >> -- >> 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/PwmizxFahYE/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> scikit-image+unsubscribe(a)googlegroups.com. >> To post to this group, send email to scikit-image(a)googlegroups.com. >> To view this discussion on the web, visit >> https://groups.google.com/d/msgid/scikit-image/CA%2BJHcKT5TijcW_Xa%3DEJt%2B… >> <https://groups.google.com/d/msgid/scikit-image/CA%2BJHcKT5TijcW_Xa%3DEJt%2B…> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > > > -- > Eyal Saiet > > Project manager > Remote sensing and in-situ measurements > > Geophysical Institute > University of Alaska Fairbanks > Fairbanks, AK 99775 > (907) 750 6555 (cell) > > ejsaiet(a)alaska.edu > > -- > 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+unsubscribe(a)googlegroups.com. > To post to this group, send email to scikit-image(a)googlegroups.com. > To view this discussion on the web, visit > https://groups.google.com/d/msgid/scikit-image/CA%2BysReD7NpvQdJgNfk_0h9b43… > <https://groups.google.com/d/msgid/scikit-image/CA%2BysReD7NpvQdJgNfk_0h9b43…> > . > > For more options, visit https://groups.google.com/d/optout. >

1 0

Re: label_boundaries problem with ndim = label_img.ndim
by Eyal Saiet Dec. 2, 2015

Dec. 2, 2015

thanks Juan, I have tried that too: 1. label_image=measure.label(g_median_bin, background=0) 2. label_image +=1 3. plt.figure() 4. plt.imshow(label_image,cmap=plt.cm.gray,interpolation='nearest',origin ='lower') 5. plt.title('label_image') 6. plt.show() 7. props=measure.regionprops(label_image,intensity_image=g_median_bin) 8. label_bound=segmentation.find_boundaries(label_image,connectivity=1, mode='thick',background=0) 9. boundaries=segmentation.mark_boundaries(g_median,label_bound) Traceback (most recent call last): File attributes_thrash.py", line 127, in <module> boundaries=segmentation.mark_boundaries(g_median,label_bound)#,color=(1,0,0), outline_color=None, mode='outer', background_label=0) File "/usr/local/lib/python3.4/dist-packages/skimage/segmentation/boundaries.py", line 213, in mark_boundaries background=background_label) File "/usr/local/lib/python3.4/dist-packages/skimage/segmentation/boundaries.py", line 156, in find_boundaries max_label = np.iinfo(label_img.dtype).max File "/usr/local/lib/python3.4/dist-packages/numpy/core/getlimits.py", line 254, in __init__ raise ValueError("Invalid integer data type.") ValueError: Invalid integer data type. Process finished with exit code 1 On Tue, Dec 1, 2015 at 10:38 PM, Juan Nunez-Iglesias <jni.soma(a)gmail.com> wrote: > Hi Eyal, > > You are passing a list of regionprops (props, in your code) to > "find_boundaries", whereas it expects an array: a single image of labeled > objects (label_img, in your code). > > Juan. > > On Wed, Dec 2, 2015 at 5:33 PM, Eyal Saiet <ejsaiet(a)alaska.edu> wrote: > >> Stefan, >> I do not understand what is not true about the label_image object. >> bellow is the code, in addition I added a plot of the label_image object. >> >> >> >> 1. label_image=measure.label(g_median_bin, background=0) >> 2. label_image +=1 >> 3. plt.figure() >> 4. plt.imshow(label_image,cmap=plt.cm.gray,interpolation='nearest', >> origin='lower') >> 5. plt.title('label_image') >> 6. plt.show() >> 7. props = measure.regionprops(label_image,intensity_image >> =g_median_bin) >> 8. label_bound=segmentation.find_boundaries(props,connectivity=1,mode= >> 'thick',background=0) >> 9. boundaries=segmentation.mark_boundaries(g_median,label_bound) >> >> >> >> >> [image: Inline image 1] >> >> On Tue, Dec 1, 2015 at 8:09 PM, Stefan van der Walt <stefanv(a)berkeley.edu >> > wrote: >> >>> On 2015-12-01 21:00:16, Arctic_python <ejsaiet(a)alaska.edu> wrote: >>> > ndim = label_img.ndim >>> > AttributeError: 'list' object has no attribute 'ndim' >>> > >>> > 1. label_image=measure.label(g_median_bin, background=0) # >>> g_median_bin >>> > is a binary image >>> >>> That does not seem to be true, given the error message. >>> >>> Stéfan >>> >>> -- >>> 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/PwmizxFahYE/unsubscribe. >>> To unsubscribe from this group and all its topics, send an email to >>> scikit-image+unsubscribe(a)googlegroups.com. >>> To post to this group, send an email to scikit-image(a)googlegroups.com. >>> To view this discussion on the web, visit >>> https://groups.google.com/d/msgid/scikit-image/87zixt8mie.fsf%40berkeley.edu >>> . >>> For more options, visit https://groups.google.com/d/optout. >>> >> >> >> >> -- >> Eyal Saiet >> >> Project manager >> Remote sensing and in-situ measurements >> >> Geophysical Institute >> University of Alaska Fairbanks >> Fairbanks, AK 99775 >> (907) 750 6555 (cell) >> >> ejsaiet(a)alaska.edu >> >> -- >> 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+unsubscribe(a)googlegroups.com. >> To post to this group, send email to scikit-image(a)googlegroups.com. >> To view this discussion on the web, visit >> https://groups.google.com/d/msgid/scikit-image/CA%2BysReAKgzF%3DJXEjz7hti8s… >> <https://groups.google.com/d/msgid/scikit-image/CA%2BysReAKgzF%3DJXEjz7hti8s…> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > -- > 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/PwmizxFahYE/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > scikit-image+unsubscribe(a)googlegroups.com. > To post to this group, send email to scikit-image(a)googlegroups.com. > To view this discussion on the web, visit > https://groups.google.com/d/msgid/scikit-image/CA%2BJHcKT5TijcW_Xa%3DEJt%2B… > <https://groups.google.com/d/msgid/scikit-image/CA%2BJHcKT5TijcW_Xa%3DEJt%2B…> > . > > For more options, visit https://groups.google.com/d/optout. > -- Eyal Saiet Project manager Remote sensing and in-situ measurements Geophysical Institute University of Alaska Fairbanks Fairbanks, AK 99775 (907) 750 6555 (cell) ejsaiet(a)alaska.edu

1 0

Re: label_boundaries problem with ndim = label_img.ndim
by Juan Nunez-Iglesias Dec. 2, 2015

Dec. 2, 2015

Hi Eyal, You are passing a list of regionprops (props, in your code) to "find_boundaries", whereas it expects an array: a single image of labeled objects (label_img, in your code). Juan. On Wed, Dec 2, 2015 at 5:33 PM, Eyal Saiet <ejsaiet(a)alaska.edu> wrote: > Stefan, > I do not understand what is not true about the label_image object. > bellow is the code, in addition I added a plot of the label_image object. > > > > 1. label_image=measure.label(g_median_bin, background=0) > 2. label_image +=1 > 3. plt.figure() > 4. plt.imshow(label_image,cmap=plt.cm.gray,interpolation='nearest', > origin='lower') > 5. plt.title('label_image') > 6. plt.show() > 7. props = measure.regionprops(label_image,intensity_image > =g_median_bin) > 8. label_bound=segmentation.find_boundaries(props,connectivity=1,mode= > 'thick',background=0) > 9. boundaries=segmentation.mark_boundaries(g_median,label_bound) > > > > > [image: Inline image 1] > > On Tue, Dec 1, 2015 at 8:09 PM, Stefan van der Walt <stefanv(a)berkeley.edu> > wrote: > >> On 2015-12-01 21:00:16, Arctic_python <ejsaiet(a)alaska.edu> wrote: >> > ndim = label_img.ndim >> > AttributeError: 'list' object has no attribute 'ndim' >> > >> > 1. label_image=measure.label(g_median_bin, background=0) # >> g_median_bin >> > is a binary image >> >> That does not seem to be true, given the error message. >> >> Stéfan >> >> -- >> 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/PwmizxFahYE/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> scikit-image+unsubscribe(a)googlegroups.com. >> To post to this group, send an email to scikit-image(a)googlegroups.com. >> To view this discussion on the web, visit >> https://groups.google.com/d/msgid/scikit-image/87zixt8mie.fsf%40berkeley.edu >> . >> For more options, visit https://groups.google.com/d/optout. >> > > > > -- > Eyal Saiet > > Project manager > Remote sensing and in-situ measurements > > Geophysical Institute > University of Alaska Fairbanks > Fairbanks, AK 99775 > (907) 750 6555 (cell) > > ejsaiet(a)alaska.edu > > -- > 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+unsubscribe(a)googlegroups.com. > To post to this group, send email to scikit-image(a)googlegroups.com. > To view this discussion on the web, visit > https://groups.google.com/d/msgid/scikit-image/CA%2BysReAKgzF%3DJXEjz7hti8s… > <https://groups.google.com/d/msgid/scikit-image/CA%2BysReAKgzF%3DJXEjz7hti8s…> > . > > For more options, visit https://groups.google.com/d/optout. >

1 0

label_boundaries problem with ndim = label_img.ndim
by Arctic_python Dec. 2, 2015

Dec. 2, 2015

Hello, I am trying to label the boundaries of an image that was segmented and labelled. I am doing something wrong because the code below returns auroura_bound=segmentation.find_boundaries(props,connectivity=1,mode='thick',background=0) File "/usr/local/lib/python3.4/dist-packages/skimage/segmentation/boundaries.py", line 148, in find_boundaries ndim = label_img.ndim AttributeError: 'list' object has no attribute 'ndim' 1. label_image=measure.label(g_median_bin, background=0) # g_median_bin is a binary image 2. label_image=label_image+1 # adjusted from -1 to 0 3. props = measure.regionprops(label_image,intensity_image=g_median_bin) # creating a regionprops object 4. label_bound=segmentation.find_boundaries(props,connectivity=1,mode= 'thick',background=0) # finding boundaries 5. boundaries=segmentation.mark_boundaries(g_median,label_bound) label_bound=segmentation.find_boundaries(props,connectivity=1,mode='thick',background=0) File "/usr/local/lib/python3.4/dist-packages/skimage/segmentation/boundaries.py", line 148, in find_boundaries ndim = label_img.ndim AttributeError: 'list' object has no attribute 'ndim' I think line 4. is redundant and not required for mark_boundary?!

3 2

Re: Strange Travis error on #1794
by Julius Bier Kirkegaard Dec. 1, 2015

Dec. 1, 2015

hmmm, same here: https://travis-ci.org/scikit-image/scikit-image/jobs/94152543 On 1 December 2015 at 21:26, Stéfan van der Walt <stefanv(a)berkeley.edu> wrote: > It looks like the docs build is breaking unrelated to #1794: > > https://travis-ci.org/scikit-image/scikit-image/jobs/94241837 > > Any ideas? > > -- > 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+unsubscribe(a)googlegroups.com. > To post to this group, send an email to scikit-image(a)googlegroups.com. > To view this discussion on the web, visit > https://groups.google.com/d/msgid/scikit-image/CABDkGQ%3DmYOXhpPovAcXmDkcEa… > . > For more options, visit https://groups.google.com/d/optout. >

1 0

Strange Travis error on #1794
by Stéfan van der Walt Dec. 1, 2015

Dec. 1, 2015

It looks like the docs build is breaking unrelated to #1794: https://travis-ci.org/scikit-image/scikit-image/jobs/94241837 Any ideas?

1 0

Pull request for LineModel3D to be used with RANSAC
by Kevin Keraudren Dec. 1, 2015

Dec. 1, 2015

Hi, I just sent a pull request for a proposed LineModel3D to be used with RANSAC: https://github.com/scikit-image/scikit-image/pull/1795 I included an example and unit tests and I would be happy to receive comments and feedback. Kind Regards, Kevin

1 0

Radius-scale matching in Determinant of Hessian (DoH) blob detector
by Iakovos Halegoua Dec. 1, 2015

Dec. 1, 2015

Hello there, I'd like to point out to the documentation authors a minor error in the blob_doh <http://scikit-image.org/docs/dev/api/skimage.feature.html#blob-doh> description. In the Notes section it is stated that: "The radius of each blob is approximately sigma.". As I tested the DoH responses of synthetic blob images, it seemed that the radius of the blob is matched to the scale according to: r = sqrt(2) * s. Could you provide me with some reference on how you calculated the matching scale to be at r = s? I guess this is nothing serious, but it was a little confusing when I first encountered it :P

3 2