Mailman 3 April 2013 - scikit-image

Re: Google Summer of Code 2013
by Umesh Sharma 15 Apr '13

15 Apr '13

@stefan Hi, Thanks for reply . I have read the modules of scikit-image and understand the modules . First i was thinking of implementing otsu's thersholding algorithm .But it is already implemented . So now I am thinking of implementing multi-level thresholding or graph cut segmentation , I have read the graph based image segmentation paper . and also started implementing it on graph (currently ) , I have installation problems (with latest source code) due to some dependencies of scipy on BLAS and LAPACK ,that i will resolve after my end semester exam(after 19th april) . and i also there is only canny edge detection algorithm for edge detection . although it is better the many other edge detection algorithm . So is it useful if there exist some more edge detection techniques ? and out of these above three which is more important to implement ( according to me since segmentation is more difficult problem so segmentation should be important ). Thank you Umesh Kumar Sharma 4th year UG student IIT Kharagpur

2 1

Re: GSoC 2013 Pre-roposal
by Raunaq Vohra 15 Apr '13

15 Apr '13

Thanks for your reply! I'm not sure if people are working on this right now, but would it be possible for me to work on an implementation of the Stroke Width Transform for text detection? I read the paper today and tried putting in a little work on it after stumbling upon it in the open issues. I was a little flustered when I realized that converting my image to dtype uint8 for the rank filter made it difficult to get clear results. But I'd really love to take this a lot further, if possible. Regards Raunaq On Sunday, April 14, 2013 9:50:50 PM UTC+5:30, Stefan van der Walt wrote: > > Hi Raunaq > > On Sun, Apr 14, 2013 at 12:53 AM, Raunaq Vohra <ronv...(a)gmail.com<javascript:> > > wrote: > >> I have never really contributed to an open-source project, so I'm very >> green (in fact, just finished with a minor bugfix, putting up a pull >> request soon), but I'm very eager to learn and I hope I can do so - in >> whatever capacity - over the next few months. At the very least, I hope to >> contribute more actively to this project and meet like-minded people. >> > > We're always very glad to have new and motivated people on board. > > My proposal for this year's GSoC is to implement image segmentation >> algorithms like the graph cut and normalized cut algorithms. I have a fair >> idea about the normalized cut method through a paper (Segmenting Point >> Sets, Yamazaki, Natarajan, Bai & Hasmann, University of California, Davis) >> and Shi & Malik's Eigenvector method and will study more about the others >> as well. >> > > There's quite a bit of interest in implementing image segmentation > algorithms, so I suggest also having a backup project idea. The ideas on > the wiki page are not the only feasible ones--so use them as guidelines, > but feel free to suggest others. > > >> I hope to find a mentor who can guide me with this. I would really >> appreciate it if someone could help with the same as I'm a little foggy >> about whom I should speak with. >> > > The mentors for this year's GSoC will be Emmanuelle Gouillart, Johannes > Schönberger, Tony Yu and myself, along with any other interested members of > the team. This mailing list is the right place to discuss project > proposals. > > Thanks for your bug-fix PR. It looks like that one was already done by > Johannes, but fortunately there are plenty of others where that came from :) > > Regards > Stéfan > >

1 0

Re: Morphological techniques
by Chintak Sheth 15 Apr '13

15 Apr '13

Hi Stefan, Thanks for the prompt reply. I'll try working on the convex_hull_objects function to start with. And go through IEEE Transactions on IP. Actually, I did go through the bugs and in particular was looking at 'Add function to resample arrays by a given factor?' ( https://github.com/scikit-image/scikit-image/issues/492), in case no one is working on it already. Chintak

1 0

Re: Morphological techniques
by Stéfan van der Walt 15 Apr '13

15 Apr '13

Hi Chintak On Mon, Apr 15, 2013 at 7:55 PM, Chintak Sheth <chintaksheth(a)gmail.com> wrote: > I am really excited about being a part of this community and have been > pretty fascinated by morphological techniques in particular. I would really > want to contribute something in this direction. Can someone please point me > to possible bugs or algorithms in this field that need implementations ? Welcome! There are several places to start, e.g. we need a convex_hull_objects function, as well as adapting rank filters to handle int16 correctly (not as 12-bit integers). Otherwise, I'd also keep an eye on IEEE Transactions on Image Processing, and what other leading image processing packages implement. There are some pretty interesting segmentation and feature detection applications of morphology too. The sky's the limit :) Cheers Stéfan P.S. There are plenty of bugs to fix too: https://github.com/scikit-image/scikit-image

1 0

Morphological techniques
by Chintak Sheth 15 Apr '13

15 Apr '13

Hi guys, I've been fiddling around with the library for some time now. And, its really brilliant to have efficient implementations of a bunch of algorithms in 1 place and seeing just a simple function call take care of the task. I am really excited about being a part of this community and have been pretty fascinated by morphological techniques in particular. I would really want to contribute something in this direction. Can someone please point me to possible bugs or algorithms in this field that need implementations ? Thanks, Chintak

1 0

Re: Errors while compiling and $make coverage
by Johannes Schönberger 15 Apr '13

15 Apr '13

Hi, It seems you do not have all the necessary files. Look at the top, where the Cython compilation takes place and the output is: non-existing path in 'skimage/morphology': '_convex_hull.c' Try to make a new, clean clone of the repository and recompile. I also recommend you to use virtualenv(-wrapper) to install packages. Johannes Schönberger Am 15.04.2013 um 02:52 schrieb Ankit Agrawal <aaaagrawal(a)gmail.com>: > Hi everyone, > > I am having some issues with compilation. This is what I did. > > 1. cloned the repo from github. > 2. for installation : $ sudo python setup.py install > > this generates the following log : http://pastebin.com/WnAtinFK > > I also noticed that the .pyc and .c files that got generated had only read-only access. > > To test further, I did > > 3. $ make coverage > > which generated this log : http://pastebin.com/LySbvnPG , probably because it could not access the .c files. Can anybody help me out by pointing if I am doing something wrong? Thank you. > > Regards, > Ankit Agrawal. > > > -- > 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. > >

3 2

Re: GSoC 2013 Pre-roposal
by Stéfan van der Walt 14 Apr '13

14 Apr '13

Hi Raunaq On Sun, Apr 14, 2013 at 12:53 AM, Raunaq Vohra <ronvohra(a)gmail.com> wrote: > I have never really contributed to an open-source project, so I'm very > green (in fact, just finished with a minor bugfix, putting up a pull > request soon), but I'm very eager to learn and I hope I can do so - in > whatever capacity - over the next few months. At the very least, I hope to > contribute more actively to this project and meet like-minded people. > We're always very glad to have new and motivated people on board. My proposal for this year's GSoC is to implement image segmentation > algorithms like the graph cut and normalized cut algorithms. I have a fair > idea about the normalized cut method through a paper (Segmenting Point > Sets, Yamazaki, Natarajan, Bai & Hasmann, University of California, Davis) > and Shi & Malik's Eigenvector method and will study more about the others > as well. > There's quite a bit of interest in implementing image segmentation algorithms, so I suggest also having a backup project idea. The ideas on the wiki page are not the only feasible ones--so use them as guidelines, but feel free to suggest others. > I hope to find a mentor who can guide me with this. I would really > appreciate it if someone could help with the same as I'm a little foggy > about whom I should speak with. > The mentors for this year's GSoC will be Emmanuelle Gouillart, Johannes Schönberger, Tony Yu and myself, along with any other interested members of the team. This mailing list is the right place to discuss project proposals. Thanks for your bug-fix PR. It looks like that one was already done by Johannes, but fortunately there are plenty of others where that came from :) Regards Stéfan

1 0

Re: Google Summer of Code 2013
by Stéfan van der Walt 14 Apr '13

14 Apr '13

Hi Umesh On Sun, Apr 14, 2013 at 10:39 AM, Umesh Sharma <usharma01(a)gmail.com> wrote: > > I have gone through the Google summer of code ideas > list of scikit-learn , I am very interested in working on implementation of > Image segmentation algorithms and also feature extraction . I have done > implemented otsu image thresholding algorithm . I know the theory of image > segmentation . So can you tell me what to do now as next step . > The first step towards a GSoC is to bond with the community, by way of contributing a few pull requests; e.g., you may fix a couple of existing bugs, or contribute a new feature. We're looking forward to seeing your work. Best regards Stéfan

1 0

Re: Histogram backprojection
by Jérôme Kieffer 13 Apr '13

13 Apr '13

On Thu, 11 Apr 2013 23:41:35 +0200 Stéfan van der Walt <stefan(a)sun.ac.za> wrote: > On Apr 11, 2013 8:49 PM, "Jérôme Kieffer" <google(a)terre-adelie.org> wrote: > > > > On Wed, 10 Apr 2013 04:13:52 -0700 (PDT) > > abid rahman <abidrahman2(a)gmail.com> wrote: > > > > > Doesn't scikit-image has a 2d convolution function like cv2.filter2D in > > > OpenCV which enables convolution of an image (or any numpy array) with > > > custom kernel ? > > > > Convolution can be done in Fourier space, especially if the kernel is > large ... I do that often, especially on nvidia GPU where FFT is for free. > > SciPy also provides fftconvolve, for what it's worth. I was surprised to figure out that fftpack (from numpy) can be faster than fftw depending on the input size... I guess it is not more than: numpy.fft.ifft2(numpy.fft.fft2(A)*numpy.fft.fft2(B).conjuguate()) > Looking forward to seeing those opencl codes! (We have to reopen the > backend discussion soon) If cuda is an option, then I have implementations based on cufft which are really efficient (but most of us could do it so trivial it is). I consider openCL is both cleaner and more promising for open source (even if less easy) Cheers, -- Jérôme Kieffer <google(a)terre-adelie.org>

1 0

Re: multi-page TIFF writer
by Stéfan van der Walt 12 Apr '13

12 Apr '13

On Fri, Apr 12, 2013 at 2:43 PM, Hadrien Mary <hadrien.mary(a)gmail.com> wrote: > You should have a look at the excellent tiffile.py: > http://www.lfd.uci.edu/~gohlke/code/tifffile.py.html We also have an I/O plugin for that, but again without the "imsave_collection" unfctionality. It would be great if someone could add that. Stéfan

1 0