Mailman 3 September 2013 - scikit-image

Re: Get border pixels of labelled region
by Juan Nunez-Iglesias Sept. 5, 2013

Sept. 5, 2013

Hi Payal, I think you're misunderstanding the purpose of "find_boundaries"... Not quite your fault since the documentation is pretty sparse there. I'll use my favourite fuzzball as an example: [image: Inline image 1] "find_boundaries" is meant to take in a segmentation, which is an image in which each region has a different integer label, like this one: [image: Inline image 2] Then, find_boundaries merely looks for pixels where the neighbouring labels are different: [image: Inline image 3] It's not doing any edge detection per se, it's just a tool for displaying segmentations when it is easier to look at edges than at regions (above). I think what you are after is edge detection? Maybe an image like this one: [image: Inline image 4] ? For that, you should directly use the edge detection algorithms, such as Sobel or Canny, without previously thresholding your image. Have a close look at the examples gallery to figure out if any are close to what you are after, and try to apply the code there to your own image: http://scikit-image.org/docs/dev/auto_examples/ Some of the more pertinent ones, I think: http://scikit-image.org/docs/dev/auto_examples/plot_canny.html http://scikit-image.org/docs/dev/auto_examples/plot_edge_filter.html http://scikit-image.org/docs/dev/auto_examples/plot_join_segmentations.html http://scikit-image.org/docs/dev/auto_examples/plot_label.html http://scikit-image.org/docs/dev/auto_examples/plot_marked_watershed.html http://scikit-image.org/docs/dev/auto_examples/plot_random_walker_segmentat… http://scikit-image.org/docs/dev/auto_examples/plot_watershed.html http://scikit-image.org/docs/dev/auto_examples/applications/plot_coins_segm… Note that in the edge detection examples, Canny/Sobel is applied to the image *before* any thresholding — unlike the example code you provided. Juan. On Wed, Sep 4, 2013 at 5:31 PM, Payal Gupta <erpayal2010(a)gmail.com> wrote: > hello everyone... > i will try this code,in this i want to do read a image then do edge > detection after that find the boundaries of image object. but > find_boundaries return me a bool array but i need a integer array. is it > any solution of this. > > > code:: > > import Image > > from skimage import segmentation as seg,io,filter > > imge = io.imread("IMG_1.jpg") > > thresh = filter.threshold_otsu(imge) > > binary = imge> thresh > > edge_canny = filter.canny(binary) > > bimg =seg.find_boundaries(binary) > > print(bimg) > > fig, (ax0, ax1) = plt.subplots(ncols=2) > > ax0.imshow(bimg,cmap=plt.cm.gray) > ax1.imshow(binary,cmap=plt.cm.gray) > > > On Tue, Sep 3, 2013 at 7:42 PM, Stéfan van der Walt <stefan(a)sun.ac.za>wrote: > >> >> On 3 Sep 2013 11:45, "Payal Gupta" <erpayal2010(a)gmail.com> wrote: >> > >> > thanks it working but i confused when i use function of skimage >> library. i want to change rgb image to a graylevel image but inbuilt >> function rgb2gray not convert gray level image. >> > reply. >> >> If you want help on this list, please provide code examples of what you >> are attempting. >> >> 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/rXRS0KT2PdU/unsubscribe. >> To unsubscribe from this group and all of its topics, send an email to >> scikit-image+unsubscribe(a)googlegroups.com. >> For more options, visit https://groups.google.com/groups/opt_out. >> > > -- > 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. > For more options, visit https://groups.google.com/groups/opt_out. >

1 0

edge_detetection
by Payal Gupta Sept. 5, 2013

Sept. 5, 2013

hello everyone... i will try this code,in this i want to do read a image then do edge detection after that find the boundaries of image object. but find_boundaries return me a bool array but i need a integer array. is it any solution of this. also edge detection not work properly. please reply me ... code:: import Image from skimage import segmentation as seg,io,filter imge = io.imread("IMG_1.jpg") thresh = filter.threshold_otsu(imge) binary = imge> thresh edge_canny = filter.canny(binary) bimg =seg.find_boundaries(binary) print(bimg) fig, (ax0, ax1) = plt.subplots(ncols=2) ax0.imshow(bimg,cmap=plt.cm.gray) ax1.imshow(binary,cmap=plt.cm.gray)

1 0

uniform_filter not working with NANs present?
by Michael Aye Sept. 5, 2013

Sept. 5, 2013

Hi! I am trying to use ndimage's uniform_filter (for a simple local mean filtering) on a map-projected image that has NAN's at the border (basically the corners where the rotated map projected image does not fit into the rectangular grid). Is there a way, maybe in skimage, to use a uniform_filter on an array that contains NANs? ndimage' version does not cope with it correctly: > arr = array([np.nan, 1,2,3,4,5,np.nan]) > arr array([ nan, 1., 2., 3., 4., 5., nan]) > nd.filters.uniform_filter(arr, 3) array([ nan, nan, nan, nan, nan, nan, nan]) Cheers, Michael

2 1

Re: Get border pixels of labelled region
by Payal Gupta Sept. 4, 2013

Sept. 4, 2013

hello everyone... i will try this code,in this i want to do read a image then do edge detection after that find the boundaries of image object. but find_boundaries return me a bool array but i need a integer array. is it any solution of this. code:: import Image from skimage import segmentation as seg,io,filter imge = io.imread("IMG_1.jpg") thresh = filter.threshold_otsu(imge) binary = imge> thresh edge_canny = filter.canny(binary) bimg =seg.find_boundaries(binary) print(bimg) fig, (ax0, ax1) = plt.subplots(ncols=2) ax0.imshow(bimg,cmap=plt.cm.gray) ax1.imshow(binary,cmap=plt.cm.gray) On Tue, Sep 3, 2013 at 7:42 PM, Stéfan van der Walt <stefan(a)sun.ac.za>wrote: > > On 3 Sep 2013 11:45, "Payal Gupta" <erpayal2010(a)gmail.com> wrote: > > > > thanks it working but i confused when i use function of skimage library. > i want to change rgb image to a graylevel image but inbuilt function > rgb2gray not convert gray level image. > > reply. > > If you want help on this list, please provide code examples of what you > are attempting. > > 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/rXRS0KT2PdU/unsubscribe. > To unsubscribe from this group and all of its topics, send an email to > scikit-image+unsubscribe(a)googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out. >

1 0

Re: Request: usage examples, demos, testimonials
by Stuart Mumford Sept. 4, 2013

Sept. 4, 2013

Hello, Do you still have the copy of that Sun warp code? Stuart On 1 Sep 2013 09:07, "Stéfan van der Walt" <stefan(a)sun.ac.za> wrote: > Hello, everyone > > I'd like to give a talk on scikit-image at the upcoming PyCon ZA [1]. > If you have any examples of the use of scikit-image in research or > industry, I'd love to hear about it. Also, impressive demo scripts > (like Stuart Mumford's sun warping), even if you can just share > snapshots, would be very valuable. > > Thanks! > Stéfan > > [1] http://za.pycon.org/ > > -- > 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. > For more options, visit https://groups.google.com/groups/opt_out. >

1 0

Re: Request: usage examples, demos, testimonials
by Jostein Bø Fløystad Sept. 3, 2013

Sept. 3, 2013

A version of random_walker which conserved memory better would be nice, sure! But I have high hopes for SLIC :) So far it has only been used in unpblished work, and unfortunately, I cannot share that work at the moment. But I will use the SART reconstruction code in a publication some time during the next few months, along with several other skimage functions. 2013/9/2 Stéfan van der Walt <stefan(a)sun.ac.za> > On Mon, Sep 2, 2013 at 9:51 PM, Jostein Bø Fløystad > <jostein.floystad(a)gmail.com> wrote: > > However, the memory requirements of the function were too high to allow > > processing the data set all in one piece, so I'm not sure I'd recommend > this > > as a showcase. > > But perhaps a good thing to put on our TODO :) What was the radon > transform code you submitted used in? > > 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+unsubscribe(a)googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out. >

1 0

Re: Get border pixels of labelled region
by Stéfan van der Walt Sept. 3, 2013

Sept. 3, 2013

On 3 Sep 2013 11:45, "Payal Gupta" <erpayal2010(a)gmail.com> wrote: > > thanks it working but i confused when i use function of skimage library. i want to change rgb image to a graylevel image but inbuilt function rgb2gray not convert gray level image. > reply. If you want help on this list, please provide code examples of what you are attempting. Stéfan

1 0

Re: Get border pixels of labelled region
by Guillaume Gay Sept. 3, 2013

Sept. 3, 2013

Ok, I responded a bit quickly yesterday, So here's the doc from find_boundaries (you can get that by running `find_boundaries?` in an ipython) Definition:find_boundaries(label_img) Docstring:Return bool array where boundaries between labeled regions are True. So you see that find_boundaries work on a labeled image, so my guess is you first need to threshold your image, and only then find the boundaries. It's not clear from what you say what exactly you want to segment out of the image, but maybe your work flow would include some colorspace conversion (for exemple RGB to HSV), then a thresholding, and then find_boundaries)... Hope this helps, Guillaume Le 03/09/2013 09:02, Payal Gupta a écrit : > . i want to find the boundary of this fig

2 1

Re: Get border pixels of labelled region
by Guillaume Gay Sept. 3, 2013

Sept. 3, 2013

You should use `io.imread('IMG_1.jpg')` instead of `Image.open`, which returns an `Image` object, not a numpy array. Le 02/09/2013 18:25, Payal Gupta a écrit : > > code: > > import Image > > from skimage import segmentation as seg,io > > img = Image.open("IMG_1.jpg" , 'r') > > bimg =seg.find_boundaries(img) > > io.imshow(bimg) > > > error: > > Traceback (most recent call last): > > File "<stdin>", line 1, in <module> > > File > "C:\Anaconda\lib\site-packages\spyderlib\widgets\externalshell\sitecustomize.py", > line 523, in runfile > > execfile(filename, namespace) > > File "C:\Users\Dell\Documents\Python Scripts\final_try.py", line 12, > in <module> > > bimg =seg.find_boundaries(img) > > File > "C:\Anaconda\lib\site-packages\skimage\segmentation\boundaries.py", > line 10, in find_boundaries > > boundaries = np.zeros(label_img.shape, dtype=np.bool) > > File "C:\Anaconda\lib\site-packages\PIL\Image.py", line 512, in > __getattr__ > > raise AttributeError(name) > > AttributeError: shape > > > > On Mon, Sep 2, 2013 at 8:27 PM, Stéfan van der Walt <stefan(a)sun.ac.za > <mailto:stefan@sun.ac.za>> wrote: > > On Mon, Sep 2, 2013 at 4:48 PM, Payal Gupta <erpayal2010(a)gmail.com > <mailto:erpayal2010@gmail.com>> wrote: > > but this syntax segmentation.find_boundaries(img) not work > > where img is an image > > Unless you send us a code snippet (e.g. on https://gist.github.com), > there is nothing we can do to help. > > 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/rXRS0KT2PdU/unsubscribe. > To unsubscribe from this group and all of its topics, send an > email to scikit-image+unsubscribe(a)googlegroups.com > <mailto:scikit-image%2Bunsubscribe@googlegroups.com>. > For more options, visit https://groups.google.com/groups/opt_out. > > > -- > 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. > For more options, visit https://groups.google.com/groups/opt_out.

2 1

Re: Request: usage examples, demos, testimonials
by Stéfan van der Walt Sept. 2, 2013

Sept. 2, 2013

On Mon, Sep 2, 2013 at 9:51 PM, Jostein Bø Fløystad <jostein.floystad(a)gmail.com> wrote: > However, the memory requirements of the function were too high to allow > processing the data set all in one piece, so I'm not sure I'd recommend this > as a showcase. But perhaps a good thing to put on our TODO :) What was the radon transform code you submitted used in? Stéfan

1 0