From elyraz at mail.com Fri May 1 13:17:52 2015 From: elyraz at mail.com (elyraz at mail.com) Date: Fri, 1 May 2015 10:17:52 -0700 (PDT) Subject: Remove grid lines and background Message-ID: Hi, I have an image (http://www.wetnewf.org/pdfs/hemocytometer.html) and I am looking for way to generate a new image that will only show the cells and omit the background and the grid lines with minimum impact on the cell dimensions. Is it possible? How can I do it? Thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: From vighneshbirodkar at gmail.com Fri May 1 15:17:18 2015 From: vighneshbirodkar at gmail.com (Vighnesh Birodkar) Date: Fri, 1 May 2015 12:17:18 -0700 (PDT) Subject: GSoC 2015 - Accepted Students Message-ID: Hello Everyone Please join in in congratulating Aman and Daniil for being a part of GSoC 2015 for scikit-image. We look forward to an awesome summer. I'd like to request Aman and Daniil to join the GSoC specific mailing list[1]. This will mainly hold project specific discussions and updates. Anyone interested in the projects can also join and share their thoughts. As part of PSF, GSoC students are required to write a blog post every 2 weeks. It would be nice if we could have an introductory blog post about each project during the community bonding period. About Daniil's project specifically, I'd like to see an overview of his face detection algorithm. Specifically, I'd like to see a pictorial representation of MB-LBP and what exactly the Viola-Jones method has patented. Please keep in mind, that these blog posts will not only be read by the mentors, but the general public as well, so they should be easily understandable and also include all the necessary details. A well documented face detection implementation would help gain a lot of traction for the project. Similarly (although I'm not clear about the specifics) if the design and strategy of the cython implementation of ndimage is well documented and explained, it will help the future maintainers enormously. For all your blog posts, keep this scenario in mind : A user googles "Face Detection / Python image filters ", and sequentially clicks each link till she/he finds what he wants. I'd also request Stefan and Ralf to chip in with their thoughts and share if they agree/disagree with my ideas about these blog posts (I'm trying to get a hang of this mentoring thing). Also, Juan can share his thoughts, for he use to always push me (and rightfully so) towards writing better last year. As a reference, see the blog feeds from last year written by Rishabh and me. You can also dig into previous years' GSoC archives to find more. Thanks Vighnesh [1] : https://groups.google.com/forum/#!forum/skimage-gsoc [2] : https://sharkysoc14.wordpress.com/ [3] : https://vcansimplify.wordpress.com/tag/gsoc/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From emmanuelle.gouillart at nsup.org Fri May 1 13:59:19 2015 From: emmanuelle.gouillart at nsup.org (Emmanuelle Gouillart) Date: Fri, 1 May 2015 19:59:19 +0200 Subject: Remove grid lines and background In-Reply-To: References: Message-ID: <20150501175919.GA1340624@phare.normalesup.org> Hi, what you want to do (separating objects from background) is called image segmentation. Take a look at the tutorial on segmentation http://scikit-image.org/docs/stable/user_guide/tutorial_segmentation.html in the user guide of scikit-image to know more about available methods. Ad for the grid lines, you can detect lines using Hough transform, then remove these regions once detected http://scikit-image.org/docs/dev/auto_examples/plot_line_hough_transform.html#example-plot-line-hough-transform-py Hope this helps Emma On Fri, May 01, 2015 at 10:17:52AM -0700, elyraz at mail.com wrote: > Hi, > I have an image (http://www.wetnewf.org/pdfs/hemocytometer.html) and I am > looking for way to generate a new image that will only show the cells and omit > the background and the grid lines with minimum impact on the cell dimensions. > Is it possible? How can I do it? > Thanks From julien.derr at gmail.com Sat May 2 08:06:05 2015 From: julien.derr at gmail.com (Julien Derr) Date: Sat, 2 May 2015 05:06:05 -0700 (PDT) Subject: problem with io.imread In-Reply-To: References: Message-ID: Thanks everyone for all the answers, I tried to desinstal reinstall with many variations. my last choice was : scikit-image==0.12dev easy_install -U scikit-image I tried your option; with plugin='matplotlib' and still get the following error message, so it looks like I have indeed a problem with a jpeg decoder, but I checked that I have the package libjpeg8 installed already. Traceback (most recent call last): File "test.py", line 17, in rgb=io.imread(cheminfichier,plugin='matplotlib') File "/usr/local/lib/python2.7/dist-packages/scikit_image-0.12dev-py2.7-linux-x86_64.egg/skimage/io/_io.py", line 100, in imread img = call_plugin('imread', fname, plugin=plugin, **plugin_args) File "/usr/local/lib/python2.7/dist-packages/scikit_image-0.12dev-py2.7-linux-x86_64.egg/skimage/io/manage_plugins.py", line 207, in call_plugin return func(*args, **kwargs) File "/usr/local/lib/python2.7/dist-packages/matplotlib/pyplot.py", line 2215, in imread return _imread(*args, **kwargs) File "/usr/local/lib/python2.7/dist-packages/matplotlib/image.py", line 1254, in imread im = pilread(fname) File "/usr/local/lib/python2.7/dist-packages/matplotlib/image.py", line 1235, in pilread return pil_to_array(image) File "/usr/local/lib/python2.7/dist-packages/matplotlib/image.py", line 1344, in pil_to_array x = toarray(im) File "/usr/local/lib/python2.7/dist-packages/matplotlib/image.py", line 1327, in toarray x_str = im.tobytes('raw', im.mode) File "/usr/local/lib/python2.7/dist-packages/PIL/Image.py", line 661, in tobytes self.load() File "/usr/local/lib/python2.7/dist-packages/PIL/ImageFile.py", line 203, in load d = Image._getdecoder(self.mode, d, a, self.decoderconfig) File "/usr/local/lib/python2.7/dist-packages/PIL/Image.py", line 419, in _getdecoder raise IOError("decoder %s not available" % decoder_name) IOError: decoder jpeg not available On Wednesday, April 29, 2015 at 6:21:21 PM UTC+2, Julien Derr wrote: > > Hi everyone, > I have a very basic problem, but I have no idea how to solve it ! > I installed scikitimage on a new computer, and I cannot manage to load an > image with the io module > > from skimage import io, works fine : > > Variable Type Data/Info > ------------------------------ > filename str ./small.jpg > io module from<...>skimage/io/__init__.pyc'> > > but when I want to load my image: I get the error > > In [5]: camera = io.imread(filename) > > ValueError: Could not load "./small.jpg" > Please see documentation at: > http://pillow.readthedocs.org/en/latest/installation.html#external-libraries > > > I went on the website where they ask to unistall PIL before installing > pillow, this is what I did, and still I get the error ... > > any ideas ? > > > thanks a lot! > > Julien > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Sat May 2 19:30:35 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Sat, 02 May 2015 16:30:35 -0700 Subject: problem with io.imread In-Reply-To: References: Message-ID: <87383ezhes.fsf@berkeley.edu> Hi Julien On 2015-05-02 05:06:05, Julien Derr wrote: > I tried your option; with plugin='matplotlib' and still get the > following error message, so it looks like I have indeed a > problem with a jpeg decoder, but I checked that I have the > package libjpeg8 installed already. Your PIL seems broken. I would completely remove PIL, make sure libjpeg and libjpeg-dev is installed, and then install Pillow instead. St?fan From warmspringwinds at gmail.com Sun May 3 14:43:22 2015 From: warmspringwinds at gmail.com (Daniil Pakhomov) Date: Sun, 3 May 2015 11:43:22 -0700 (PDT) Subject: GSoC 2015 - Accepted Students In-Reply-To: References: Message-ID: <660d6eee-9940-470e-8519-a9bee47366b4@googlegroups.com> Thanks, Vignesh :) And I am also happy to see you as a mentor. Yes, I think it's a good idea. I will write it. ???????, 1 ??? 2015 ?., 21:17:19 UTC+2 ???????????? Vighnesh Birodkar ???????: > > Hello Everyone > > Please join in in congratulating Aman and Daniil for being a part of GSoC > 2015 for scikit-image. We look forward to an awesome summer. > > I'd like to request Aman and Daniil to join the GSoC specific mailing > list[1]. This will mainly hold project specific discussions and updates. > Anyone interested in the projects can also join and share their thoughts. > > As part of PSF, GSoC students are required to write a blog post every 2 > weeks. It would be nice if we could have an introductory blog post about > each project during the community bonding period. About Daniil's project > specifically, I'd like to see an overview of his face detection algorithm. > Specifically, I'd like to see a pictorial representation of MB-LBP and what > exactly the Viola-Jones method has patented. > > Please keep in mind, that these blog posts will not only be read by the > mentors, but the general public as well, so they should be easily > understandable and also include all the necessary details. A well > documented face detection implementation would help gain a lot of traction > for the project. Similarly (although I'm not clear about the specifics) if > the design and strategy of the cython implementation of ndimage is well > documented and explained, it will help the future maintainers enormously. > > For all your blog posts, keep this scenario in mind : A user googles "Face > Detection / Python image filters ", and sequentially clicks each link till > she/he finds what he wants. > > I'd also request Stefan and Ralf to chip in with their thoughts and share > if they agree/disagree with my ideas about these blog posts (I'm trying to > get a hang of this mentoring thing). Also, Juan can share his thoughts, for > he use to always push me (and rightfully so) towards writing better last > year. > > As a reference, see the blog feeds from last year written by Rishabh and > me. You can also dig into previous years' GSoC archives to find more. > > Thanks > Vighnesh > > [1] : https://groups.google.com/forum/#!forum/skimage-gsoc > [2] : https://sharkysoc14.wordpress.com/ > [3] : https://vcansimplify.wordpress.com/tag/gsoc/ > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Sun May 3 16:03:39 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Sun, 03 May 2015 13:03:39 -0700 Subject: GSoC 2015 - Accepted Students In-Reply-To: References: Message-ID: <87vbg9xwbo.fsf@berkeley.edu> On 2015-05-01 12:17:18, Vighnesh Birodkar wrote: > I'd also request Stefan and Ralf to chip in with their thoughts > and share if they agree/disagree with my ideas about these blog > posts (I'm trying to get a hang of this mentoring thing). Also, > Juan can share his thoughts, for he use to always push me (and > rightfully so) towards writing better last year. For me, it is important that we have frequent (daily) interaction during the community bonding period. So, let's get going on writing those blogs, filing pull requests, reviewing existing PRs, and chatting on gitter! We have a killer team, both mentor- and student-wise this year--I'm excited! St?fan From emmanuelle.gouillart at nsup.org Sun May 3 09:57:53 2015 From: emmanuelle.gouillart at nsup.org (Emmanuelle Gouillart) Date: Sun, 3 May 2015 15:57:53 +0200 Subject: Euroscipy deadline extended Message-ID: <20150503135753.GA2146626@phare.normalesup.org> Hi scikit-image folks, the deadline for submitting abstracts for talks and posters has been extended for Euroscipy 2015, to May 15. Euroscipy 2015 will be in Cambridge UK this year, from August 26 to 29 (see https://www.euroscipy.org/2015/). If you're using scikit-image for your work and you would like to share with the community the tools you develop, the usage patterns that you find interesting, or to illustrate how scikit-image can be used for applications, Euroscipy is a great opportunity! Euroscipy has always been a great conference, with a welcoming and enthusiastic atmosphere. Tutorials sessions are organised for attendees of different levels, from beginner to more advanced Pythonistas. Another nice thing of the conference is the mixture of people from academic labs, industry and start-ups: it's a great place to meet interesting people that are using the same tools, in different backgrounds and applications. This year, Juan and myself are submitting an intermediate-level tutorial about scikit-image. We cannot guarantee yet that it will be accepted, of course, but let's keep fingers crossed :-). We'll also organize a sprint on scikit-image (probably on Sunday 30, after the conference). So, please consider submitting an abstract, or just attending the conference! Also spread the word around you (you may for example retweet https://twitter.com/EGouillart/status/594073076452696066) ! Cheers, Emma From stefanv at berkeley.edu Mon May 4 02:54:33 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Sun, 03 May 2015 23:54:33 -0700 Subject: Remove grid lines and background In-Reply-To: References: Message-ID: <87a8xkygra.fsf@berkeley.edu> On 2015-05-01 10:17:52, elyraz at mail.com wrote: > I have an image (http://www.wetnewf.org/pdfs/hemocytometer.html) > and I am looking for way to generate a new image that will only > show the cells and omit the background and the grid lines with > minimum impact on the cell dimensions. In your case, the lines are so regular that you can simply take the column and row sums and see where they peak. An example is shown here (sorry, the notebook output was stripped, so you'll have to execute it to see the actual images): http://nbviewer.ipython.org/github/scikit-image/skimage-tutorials/blob/master/lectures/adv2_microarray.ipynb St?fan From julien.derr at gmail.com Mon May 4 11:41:28 2015 From: julien.derr at gmail.com (Julien Derr) Date: Mon, 4 May 2015 08:41:28 -0700 (PDT) Subject: problem with io.imread In-Reply-To: <87383ezhes.fsf@berkeley.edu> References: <87383ezhes.fsf@berkeley.edu> Message-ID: thanks a lot ! On Sunday, May 3, 2015 at 1:30:39 AM UTC+2, stefanv wrote: > > Hi Julien > > On 2015-05-02 05:06:05, Julien Derr > > wrote: > > I tried your option; with plugin='matplotlib' and still get the > > following error message, so it looks like I have indeed a > > problem with a jpeg decoder, but I checked that I have the > > package libjpeg8 installed already. > > Your PIL seems broken. I would completely remove PIL, make sure > libjpeg and libjpeg-dev is installed, and then install Pillow > instead. > > St?fan > -------------- next part -------------- An HTML attachment was scrubbed... URL: From fboulogne at sciunto.org Mon May 4 12:01:57 2015 From: fboulogne at sciunto.org (=?UTF-8?B?RnJhbsOnb2lzIEJvdWxvZ25l?=) Date: Mon, 04 May 2015 12:01:57 -0400 Subject: problem with measure.label In-Reply-To: References: Message-ID: <554797F5.4080901@sciunto.org> Le 04/05/2015 11:45, Julien Derr a ?crit : > Hi everyone, it is me again ... > I got this error with the label function > TypeError: label() got an unexpected keyword argument 'connectivity' > > but I used it exactly like written in the help : > > cc= label(binary2,connectivity=2) > > I thought it was maybe a problem of version, but it is written in my > own help when I type ?label, > so what could it be ? ideas? > > Hi Julien, I can't reproduce with label(np.zeros(5), connectivity=1). Does this line works for you? Could you provide a minimal example please? Best, -- Fran?ois Boulogne. http://www.sciunto.org GPG: 32D5F22F From stefanv at berkeley.edu Mon May 4 16:23:31 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Mon, 04 May 2015 13:23:31 -0700 Subject: GSoC 2015 - Accepted Students In-Reply-To: References: <87vbg9xwbo.fsf@berkeley.edu> Message-ID: <876188xfb0.fsf@berkeley.edu> On 2015-05-04 12:57:59, ?????? ??????? wrote: > Do you know whether it's possible to change my blog address? I > specified one address in my application and now I want to use > another one. > > Just want to make it using markdown on github and make it so > that anyone can push changes in case they find a mistake. If you create a repository myusername.github.io, you can publish a static blog there using Pelican. Or have a look at something like http://jekyllbootstrap.com/ St?fan From julien.derr at gmail.com Mon May 4 11:45:48 2015 From: julien.derr at gmail.com (Julien Derr) Date: Mon, 4 May 2015 17:45:48 +0200 Subject: problem with measure.label Message-ID: Hi everyone, it is me again ... I got this error with the label function TypeError: label() got an unexpected keyword argument 'connectivity' but I used it exactly like written in the help : cc= label(binary2,connectivity=2) I thought it was maybe a problem of version, but it is written in my own help when I type ?label, so what could it be ? ideas? thanks again! Julien -------------- next part -------------- An HTML attachment was scrubbed... URL: From emmanuelle.gouillart at nsup.org Mon May 4 11:55:14 2015 From: emmanuelle.gouillart at nsup.org (Emmanuelle Gouillart) Date: Mon, 4 May 2015 17:55:14 +0200 Subject: problem with io.imread In-Reply-To: References: <87383ezhes.fsf@berkeley.edu> Message-ID: <20150504155514.GA2583429@phare.normalesup.org> did it work? On Mon, May 04, 2015 at 08:41:28AM -0700, Julien Derr wrote: > thanks a lot ! > On Sunday, May 3, 2015 at 1:30:39 AM UTC+2, stefanv wrote: > Hi Julien > On 2015-05-02 05:06:05, Julien Derr wrote: > > I tried your option; with plugin='matplotlib' and still get the > > following error message, so it looks like I have indeed a > > problem with a jpeg decoder, but I checked that I have the > > package libjpeg8 installed already. > Your PIL seems broken. I would completely remove PIL, make sure > libjpeg and libjpeg-dev is installed, and then install Pillow > instead. > St??fan From warmspringwinds at gmail.com Mon May 4 15:57:59 2015 From: warmspringwinds at gmail.com (=?UTF-8?B?0JTQsNC90LjQuNC7INCf0LDRhdC+0LzQvtCy?=) Date: Mon, 4 May 2015 21:57:59 +0200 Subject: GSoC 2015 - Accepted Students In-Reply-To: <87vbg9xwbo.fsf@berkeley.edu> References: <87vbg9xwbo.fsf@berkeley.edu> Message-ID: Awesome :) Do you know whether it's possible to change my blog address? I specified one address in my application and now I want to use another one. Just want to make it using markdown on github and make it so that anyone can push changes in case they find a mistake. 2015-05-03 22:03 GMT+02:00 Stefan van der Walt : > On 2015-05-01 12:17:18, Vighnesh Birodkar > wrote: > >> I'd also request Stefan and Ralf to chip in with their thoughts and >> share if they agree/disagree with my ideas about these blog posts (I'm >> trying to get a hang of this mentoring thing). Also, Juan can share his >> thoughts, for he use to always push me (and rightfully so) towards writing >> better last year. >> > > For me, it is important that we have frequent (daily) interaction during > the community bonding period. So, let's get going on writing those blogs, > filing pull requests, reviewing existing PRs, and chatting on gitter! > > We have a killer team, both mentor- and student-wise this year--I'm > excited! > > St?fan > > > -- > 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/M6rJahjn610/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > scikit-image+unsubscribe at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ralf.gommers at gmail.com Mon May 4 16:09:10 2015 From: ralf.gommers at gmail.com (Ralf Gommers) Date: Mon, 4 May 2015 22:09:10 +0200 Subject: GSoC 2015 - Accepted Students In-Reply-To: <87vbg9xwbo.fsf@berkeley.edu> References: <87vbg9xwbo.fsf@berkeley.edu> Message-ID: On Sun, May 3, 2015 at 10:03 PM, Stefan van der Walt wrote: > On 2015-05-01 12:17:18, Vighnesh Birodkar > wrote: > >> I'd also request Stefan and Ralf to chip in with their thoughts and >> share if they agree/disagree with my ideas about these blog posts (I'm >> trying to get a hang of this mentoring thing). > > What you wrote sounds good to me. Somehow it's difficult for students to get around to writing blog posts, maybe they're afraid that their work sounds simple or would like to only write a post when "feature X" is done. There's no need though - just sharing what you have accomplished and what's next is enough, very helpful for the community to see what you're doing. And usually generates at least some feedback. Here's the best example I've seen for Scipy in the last few years: http://cwl.cx/2013/index.html > Also, Juan can share his thoughts, for he use to always push me (and >> rightfully so) towards writing better last year. >> > > For me, it is important that we have frequent (daily) interaction during > the community bonding period. So, let's get going on writing those blogs, > filing pull requests, reviewing existing PRs, and chatting on gitter! > > We have a killer team, both mentor- and student-wise this year--I'm > excited! > Couldn't agree more. I'm looking forward to this summer. Cheers, Ralf > St?fan > > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sdochengxu1993 at gmail.com Tue May 5 04:37:29 2015 From: sdochengxu1993 at gmail.com (xu cheng) Date: Tue, 5 May 2015 01:37:29 -0700 (PDT) Subject: hough ellipse fit inaccurate? In-Reply-To: <762d8ac0-1081-44aa-9420-af88f4cc590d@googlegroups.com> References: <8b232a1c-052c-44fb-8ed6-5d8bd5380761@googlegroups.com> <497ca5a4-2874-4650-8cde-bd160f1125d0@googlegroups.com> <762d8ac0-1081-44aa-9420-af88f4cc590d@googlegroups.com> Message-ID: <2303e4b4-0d2d-4455-a00b-6f83138d4a5e@googlegroups.com> Hi Arno, Using the image you provided here, I fit the black pupil center using a ellipse estimation algorithm. The result is 295.3148 along the horizontal direction from left to right 106.7683 along the vertical direction from the top to down. Since I don't know the exact pupil center for this image, I?m not sure the result is good or not. Best wishes! Xu Cheng ? 2015?3?5???? UTC+8??7:09:34?Arno Dietz??? > > Ok sorry, maybe I need to explain my project better. > I have a Autodesk Maya Model of a simplified eyeball with the pupil (see > image). I render the images of the eyeball from different angles and try to > detect the pupil center as accurate as possible. Since I know the geometry > and rotation of the > eyeball, I can calculate the mapping of the real center of the pupil on my > virtual maya camera sensor. I proved the validity > of my calculation > with several other methods of ellipse center detection (center of mass, > distance transform, opencv ellipse fit, starbust) where I get errors > between my calculation and measurement with sometimes less than 0.02 pixels. > But nevertheless I > want to try the hough ellipse approach, because it may be more robust > against noise or other errors I want to simulate later. And it seems so far > only the hough ellipse approach is quite inaccurate so I was wondering why. > I think the reason could be the ellipse center detection has only a half > pixel accuracy while my other approaches have sub-pixel accuracy. > > I think posting the calculation code would be too much, but I am quite > sure the calculation is right. > > Kind regards > Arno > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jni.soma at gmail.com Tue May 5 03:20:30 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Tue, 5 May 2015 17:20:30 +1000 Subject: GSoC 2015 - Accepted Students In-Reply-To: <876188xfb0.fsf@berkeley.edu> References: <87vbg9xwbo.fsf@berkeley.edu> <876188xfb0.fsf@berkeley.edu> Message-ID: Hi everyone, Thanks for the name drop Vighnesh! =) Welcome Aman and Daniil! I'm really looking forward to your contributions. Aman, as someone who's struggled, and failed, with the C backend of ndimage, I'm particularly interested in your project! I'm looking forward to seeing your work! As to writing, yes, good writing is a pet peeve of mine. It's tremendously important, way more than hackers like us naturally realise. And it's something that you can always improve. I'll post here something that someone sent me recently as a critique on my writing: http://designopen.org/articles/import-designers/ Read, in particular, the section under "The value of your project". As you write your blog posts, don't just write a simple log of your activities. Try to express the problems you faced and how you solved them, no matter how trivial you think they are (e.g. "I stupidly forgot that dictionary keys are unordered and it caused this bug but now it's fixed through this PR that uses OrderedDict from the standard library"). And try to express them in as broad a way as possible, for any programmer, not just people who are familiar with SciPy internals (a rather small audience!). See you around! Juan. On Tue, May 5, 2015 at 6:23 AM, Stefan van der Walt wrote: > On 2015-05-04 12:57:59, ?????? ??????? wrote: > >> Do you know whether it's possible to change my blog address? I specified >> one address in my application and now I want to use another one. >> >> Just want to make it using markdown on github and make it so that anyone >> can push changes in case they find a mistake. >> > > If you create a repository myusername.github.io, you can publish a static > blog there using Pelican. Or have a look at something like > > http://jekyllbootstrap.com/ > > St?fan > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From vighneshbirodkar at gmail.com Wed May 6 01:30:12 2015 From: vighneshbirodkar at gmail.com (Vighnesh Birodkar) Date: Tue, 5 May 2015 22:30:12 -0700 (PDT) Subject: GSoC 2015 - Accepted Students In-Reply-To: References: Message-ID: <8472c102-e280-45dc-a7a8-f4aaf8eb2489@googlegroups.com> BTW, Wordpress[1] also supports markdown, that's what I end up using the most. [1] : https://en.support.wordpress.com/markdown/ On Saturday, May 2, 2015 at 12:47:19 AM UTC+5:30, Vighnesh Birodkar wrote: > > Hello Everyone > > Please join in in congratulating Aman and Daniil for being a part of GSoC > 2015 for scikit-image. We look forward to an awesome summer. > > I'd like to request Aman and Daniil to join the GSoC specific mailing > list[1]. This will mainly hold project specific discussions and updates. > Anyone interested in the projects can also join and share their thoughts. > > As part of PSF, GSoC students are required to write a blog post every 2 > weeks. It would be nice if we could have an introductory blog post about > each project during the community bonding period. About Daniil's project > specifically, I'd like to see an overview of his face detection algorithm. > Specifically, I'd like to see a pictorial representation of MB-LBP and what > exactly the Viola-Jones method has patented. > > Please keep in mind, that these blog posts will not only be read by the > mentors, but the general public as well, so they should be easily > understandable and also include all the necessary details. A well > documented face detection implementation would help gain a lot of traction > for the project. Similarly (although I'm not clear about the specifics) if > the design and strategy of the cython implementation of ndimage is well > documented and explained, it will help the future maintainers enormously. > > For all your blog posts, keep this scenario in mind : A user googles "Face > Detection / Python image filters ", and sequentially clicks each link till > she/he finds what he wants. > > I'd also request Stefan and Ralf to chip in with their thoughts and share > if they agree/disagree with my ideas about these blog posts (I'm trying to > get a hang of this mentoring thing). Also, Juan can share his thoughts, for > he use to always push me (and rightfully so) towards writing better last > year. > > As a reference, see the blog feeds from last year written by Rishabh and > me. You can also dig into previous years' GSoC archives to find more. > > Thanks > Vighnesh > > [1] : https://groups.google.com/forum/#!forum/skimage-gsoc > [2] : https://sharkysoc14.wordpress.com/ > [3] : https://vcansimplify.wordpress.com/tag/gsoc/ > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Wed May 6 01:50:08 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Tue, 05 May 2015 22:50:08 -0700 Subject: GSoC 2015 - Accepted Students In-Reply-To: References: <87vbg9xwbo.fsf@berkeley.edu> <876188xfb0.fsf@berkeley.edu> Message-ID: <878ud2w8z3.fsf@berkeley.edu> Hi Aman On 2015-05-05 20:34:42, Aman SIngh wrote: > I have changed my blog address to acsgsoc15.wordpress.com If you > can see it and give some suggestions, that would be very > helpful. For future posts I will surely keep in mind the > suggestions on blog writing. I've updated the link on the wiki: https://github.com/scikit-image/scikit-image/wiki/GSoC-2015 Thanks for sharing your first blog post! I think it would help to clarify the difference between variables with types and without types. To say that the [] operator is slow in Cython is misleading, because mostly you will only use that operator once you've defined the type of the array. Avoid phrases like "these work best", and try to specify what you mean. For example, instead of saying "In NumPy arrays modes can also be specified. Like ?c? or ?fortran? types. These work best when according to our mode of iteration on arrays(row or column wise). ?c? mode works best when the iteration is row wise and ?fortran? for column wise. " try something like "The memory layout of a NumPy array can be either Fortran-order (elements unpacked along columns) or C-order (elements unpacked along rows). It is best to access the elements of the array in the same order that they appear in memory, in order to make efficient use of caching." As always, the standard disclaimer: feedback is given (on blogs, PRs and the like) not because we enjoy being critical, but because we believe that we all learn from one another. You'll soon enough get the chance to give feedback on one of my PRs, which are seldom anywhere near perfect! Regards St?fan From ralf.gommers at gmail.com Wed May 6 02:12:33 2015 From: ralf.gommers at gmail.com (Ralf Gommers) Date: Wed, 6 May 2015 08:12:33 +0200 Subject: GSoC 2015 - Accepted Students In-Reply-To: <878ud2w8z3.fsf@berkeley.edu> References: <87vbg9xwbo.fsf@berkeley.edu> <876188xfb0.fsf@berkeley.edu> <878ud2w8z3.fsf@berkeley.edu> Message-ID: On Wed, May 6, 2015 at 7:50 AM, Stefan van der Walt wrote: > Hi Aman > > On 2015-05-05 20:34:42, Aman SIngh wrote: > >> I have changed my blog address to acsgsoc15.wordpress.com If you can see >> it and give some suggestions, that would be very helpful. For future posts >> I will surely keep in mind the suggestions on blog writing. >> > > I've updated the link on the wiki: > > https://github.com/scikit-image/scikit-image/wiki/GSoC-2015 > Hi Aman, what is the address to use for RSS feed? The PSF admins need that, and I'd also like to add it to planet.scipy.org Ralf -------------- next part -------------- An HTML attachment was scrubbed... URL: From ug201310004 at iitj.ac.in Tue May 5 23:34:42 2015 From: ug201310004 at iitj.ac.in (Aman SIngh) Date: Wed, 6 May 2015 09:04:42 +0530 Subject: GSoC 2015 - Accepted Students In-Reply-To: References: <87vbg9xwbo.fsf@berkeley.edu> <876188xfb0.fsf@berkeley.edu> Message-ID: Hi everyone, Firstly thanks for the welcome :) I have changed my blog address to acsgsoc15.wordpress.com If you can see it and give some suggestions, that would be very helpful. For future posts I will surely keep in mind the suggestions on blog writing. @Juan Thank-you. I will try my best to come up with your expectations. I am looking forward to working with you all...!!! Regards, Aman On Tue, May 5, 2015 at 12:50 PM, Juan Nunez-Iglesias wrote: > Hi everyone, > > Thanks for the name drop Vighnesh! =) > > Welcome Aman and Daniil! I'm really looking forward to your contributions. > > Aman, as someone who's struggled, and failed, with the C backend of > ndimage, I'm particularly interested in your project! I'm looking forward > to seeing your work! > > As to writing, yes, good writing is a pet peeve of mine. It's tremendously > important, way more than hackers like us naturally realise. And it's > something that you can always improve. I'll post here something that > someone sent me recently as a critique on my writing: > http://designopen.org/articles/import-designers/ > > Read, in particular, the section under "The value of your project". As you > write your blog posts, don't just write a simple log of your activities. > Try to express the problems you faced and how you solved them, no matter > how trivial you think they are (e.g. "I stupidly forgot that dictionary > keys are unordered and it caused this bug but now it's fixed through this > PR that uses OrderedDict from the standard library"). And try to express > them in as broad a way as possible, for any programmer, not just people who > are familiar with SciPy internals (a rather small audience!). > > See you around! > > Juan. > > On Tue, May 5, 2015 at 6:23 AM, Stefan van der Walt > wrote: > >> On 2015-05-04 12:57:59, ?????? ??????? wrote: >> >>> Do you know whether it's possible to change my blog address? I >>> specified one address in my application and now I want to use another one. >>> >>> Just want to make it using markdown on github and make it so that anyone >>> can push changes in case they find a mistake. >>> >> >> If you create a repository myusername.github.io, you can publish a >> static blog there using Pelican. Or have a look at something like >> >> http://jekyllbootstrap.com/ >> >> St?fan >> >> -- >> 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 at googlegroups.com. >> For more options, visit https://groups.google.com/d/optout. >> > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- Aman Singh Coordinator Programming Club Sophomore, Computer Science and Technology Indian Institute of Technology, Jodhpur -------------- next part -------------- An HTML attachment was scrubbed... URL: From jni.soma at gmail.com Wed May 6 02:05:20 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Wed, 6 May 2015 16:05:20 +1000 Subject: GSoC 2015 - Accepted Students In-Reply-To: <878ud2w8z3.fsf@berkeley.edu> References: <87vbg9xwbo.fsf@berkeley.edu> <876188xfb0.fsf@berkeley.edu> <878ud2w8z3.fsf@berkeley.edu> Message-ID: Thanks Aman, I second St?fan's comments, and I also want to recommend that you proofread your posts before publishing them. There's quite a few typos in there, e.g. speeed and spcifying. Typos are no big deal, they happen to all of us, but they give the impression of carelessness. Everyone has a different system, but this is mine: I personally never publish as soon as I finish writing. I finish it, save it, and sleep at least one night before reading it again. If I spot any typos, they get fixed and I postpone for another day! It's slower but it makes a huge difference to the final quality. Reading five times in a row doesn't help because we become blind to existing errors. Since these posts are up forever (we hope), I think it's worth it! Juan. On Wed, May 6, 2015 at 3:50 PM, Stefan van der Walt wrote: > Hi Aman > > On 2015-05-05 20:34:42, Aman SIngh wrote: > >> I have changed my blog address to acsgsoc15.wordpress.com If you can see >> it and give some suggestions, that would be very helpful. For future posts >> I will surely keep in mind the suggestions on blog writing. >> > > I've updated the link on the wiki: > > https://github.com/scikit-image/scikit-image/wiki/GSoC-2015 > > Thanks for sharing your first blog post! > > I think it would help to clarify the difference between variables with > types and without types. To say that the [] operator is slow in Cython is > misleading, because mostly you will only use that operator once you've > defined the type of the array. > > Avoid phrases like "these work best", and try to specify what you mean. > For example, instead of saying > > "In NumPy arrays modes can also be specified. Like 'c' or 'fortran' types. > These work best when according to our mode of iteration on arrays(row or > column wise). 'c' mode works best when the iteration is row wise and > 'fortran' for column wise. " > > try something like > > "The memory layout of a NumPy array can be either Fortran-order (elements > unpacked along columns) or C-order (elements unpacked along rows). It is > best to access the elements of the array in the same order that they appear > in memory, in order to make efficient use of caching." > > As always, the standard disclaimer: feedback is given (on blogs, PRs and > the like) not because we enjoy being critical, but because we believe that > we all learn from one another. You'll soon enough get the chance to give > feedback on one of my PRs, which are seldom anywhere near perfect! > > Regards > > St?fan > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ug201310004 at iitj.ac.in Wed May 6 12:15:13 2015 From: ug201310004 at iitj.ac.in (Aman SIngh) Date: Wed, 6 May 2015 21:45:13 +0530 Subject: GSoC 2015 - Accepted Students In-Reply-To: References: <87vbg9xwbo.fsf@berkeley.edu> <876188xfb0.fsf@berkeley.edu> <878ud2w8z3.fsf@berkeley.edu> Message-ID: Hi Everyone, @Juan and @Stefan Thanks for the suggestions. I will keep them in mind for future posts and correct the typos in previous blogs. @Ralf RSS feed of the blog is acsgsoc15.wordpress.com /feed/ Aman On Wed, May 6, 2015 at 11:42 AM, Ralf Gommers wrote: > > > On Wed, May 6, 2015 at 7:50 AM, Stefan van der Walt > wrote: > >> Hi Aman >> >> On 2015-05-05 20:34:42, Aman SIngh wrote: >> >>> I have changed my blog address to acsgsoc15.wordpress.com If you can >>> see it and give some suggestions, that would be very helpful. For future >>> posts I will surely keep in mind the suggestions on blog writing. >>> >> >> I've updated the link on the wiki: >> >> https://github.com/scikit-image/scikit-image/wiki/GSoC-2015 >> > > Hi Aman, what is the address to use for RSS feed? The PSF admins need > that, and I'd also like to add it to planet.scipy.org > > Ralf > > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -- Aman Singh Coordinator Programming Club Sophomore, Computer Science and Technology Indian Institute of Technology, Jodhpur -------------- next part -------------- An HTML attachment was scrubbed... URL: From iacopo.guarneri at gmail.com Thu May 7 09:53:29 2015 From: iacopo.guarneri at gmail.com (Iacopo Guarneri) Date: Thu, 7 May 2015 06:53:29 -0700 (PDT) Subject: What classes to use in scikit Message-ID: Hello everyone! I'm trying to learn to use scikit learn Correct me if I'm wrong, from what I understand (much easier) scikit is used in this way: 1) instantiate the class for the type of learning / prediction (eg: svm, IsotonicRegression, LinearRegression etc ...) 2) launch class.fit to learn 3) launch class.predict to predict The biggest problem I have at the point 1, I understand when to use supervised learning (classification or regression) and unsupervised, but do not know which class to instantiate. I do not have a project in mind, I am interested for educational purposes and in order to solve this problem I would like to know: 1) you can advise me the best 3 classes for the 3 types of learning ( unsupervised (classification, regression) and unsupervised)? 2) you can recommend a page which explains the various classes and the cases in which are to be used? thanks :) -------------- next part -------------- An HTML attachment was scrubbed... URL: From jsch at demuc.de Thu May 7 09:26:58 2015 From: jsch at demuc.de (=?utf-8?Q?Johannes_Sch=C3=B6nberger?=) Date: Thu, 7 May 2015 09:26:58 -0400 Subject: No fixed-width font for homepage? In-Reply-To: References: Message-ID: <244DFB55-EA85-48CD-9256-84D3BA2CDF4F@demuc.de> https://github.com/scikit-image/scikit-image-web/pull/26 > On May 7, 2015, at 8:04 AM, Juan Nunez-Iglesias wrote: > > Here's a screengrab of rendering on my browser: > > > > -- > 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 at googlegroups.com . > For more options, visit https://groups.google.com/d/optout . -------------- next part -------------- An HTML attachment was scrubbed... URL: From fboulogne at sciunto.org Thu May 7 10:19:33 2015 From: fboulogne at sciunto.org (=?UTF-8?B?RnJhbsOnb2lzIEJvdWxvZ25l?=) Date: Thu, 07 May 2015 10:19:33 -0400 Subject: What classes to use in scikit In-Reply-To: References: Message-ID: <554B7475.2010809@sciunto.org> Hi, > I'm trying to learn to use scikit learn > Correct me if I'm wrong, from what I understand (much easier) scikit > is used in this way: > You are on the scikit-*Image* mailing list :) Best, -- Fran?ois Boulogne. http://www.sciunto.org GPG: 32D5F22F From jni.soma at gmail.com Thu May 7 08:04:43 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Thu, 7 May 2015 22:04:43 +1000 Subject: No fixed-width font for homepage? Message-ID: Here's a screengrab of rendering on my browser: [image: Inline image 1] -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screen Shot 2015-05-07 at 10.02.51 pm.png Type: image/png Size: 63661 bytes Desc: not available URL: From stefanv at berkeley.edu Fri May 8 04:01:17 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Fri, 08 May 2015 01:01:17 -0700 Subject: No fixed-width font for homepage? In-Reply-To: References: <244DFB55-EA85-48CD-9256-84D3BA2CDF4F@demuc.de> Message-ID: <878uczqz02.fsf@berkeley.edu> On 2015-05-07 20:46:05, Juan Nunez-Iglesias wrote: > Thanks Johannes! > > Incidentally, there's some much nicer monospace fonts out > there... Off the top of my head: Liberation Mono (Fedora's free > font), Inconsolata, Adobe Source Code Pro. Any interest in using > one of those? My browser already uses Dejavu Sans Mono by default--but PRs are welcome! St?fan From jni.soma at gmail.com Thu May 7 23:46:05 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Fri, 8 May 2015 13:46:05 +1000 Subject: No fixed-width font for homepage? In-Reply-To: <244DFB55-EA85-48CD-9256-84D3BA2CDF4F@demuc.de> References: <244DFB55-EA85-48CD-9256-84D3BA2CDF4F@demuc.de> Message-ID: Thanks Johannes! Incidentally, there's some much nicer monospace fonts out there... Off the top of my head: Liberation Mono (Fedora's free font), Inconsolata, Adobe Source Code Pro. Any interest in using one of those? Juan. On Thu, May 7, 2015 at 11:26 PM, Johannes Sch?nberger wrote: > https://github.com/scikit-image/scikit-image-web/pull/26 > > On May 7, 2015, at 8:04 AM, Juan Nunez-Iglesias > wrote: > > Here's a screengrab of rendering on my browser: > > > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joelfrederico at gmail.com Tue May 12 03:26:12 2015 From: joelfrederico at gmail.com (Joel Frederico) Date: Tue, 12 May 2015 00:26:12 -0700 (PDT) Subject: Regionprops Clarification Message-ID: http://scikit-image.org/docs/dev/api/skimage.measure.html#regionprops If we look at the documentation, the attribute weighted_moments_central uses the region's centroid to subtract when calculating the moments. Shouldn't it use the region's weighted_centroid values? Does it and it just doesn't say that? I'm a bit confused. -------------- next part -------------- An HTML attachment was scrubbed... URL: From agaldran at gmail.com Tue May 12 18:55:45 2015 From: agaldran at gmail.com (=?UTF-8?Q?Adri=C3=A1n_Galdr=C3=A1n?=) Date: Tue, 12 May 2015 15:55:45 -0700 (PDT) Subject: Fitted Ellipse and its major axis Message-ID: <288049b5-06dd-48a7-99d1-a3702b5c33c6@googlegroups.com> Good night! I have been trying to fit an ellipse to the border of a roughly elliptic region. The ellipse fits nicely, but whenever I try to plot the major axis, it fails. Of note, the approach below works well for phi \in [pi/4, 3pi/4], but not on other semi-cuadrants. I think the problem lies on the output of xc, yc, a, b, alpha = ellipse.params. Here, xc and yc are the coordinates of the center of the ellipse, a and b are the length of major and minor axis, and alpha is (I guess) the angle that the major axis makes against the horizontal axis. Does anyone have a hint for handling properly that angle, so that the code below works under any orientation of the ellipse? Many thanks! from matplotlib import pyplot as plt import numpy as np import skimage.io as io from skimage.measure import EllipseModel from skimage import feature from skimage.transform import rotate bin_im = io.imread( 'https://www.dropbox.com/s/fantqj8x4mbbsxf/bin_mask.png?dl=0') ##################################################### #This routine fails for ellipses oriented in [0,pi/4] # bin_im = rotate(bin_im, 0, order=1) # bin_im = rotate(bin_im, 30, order=1) #It also seems to fail for [3*pi/4 ,pi] #bin_im = rotate(bin_im, 140, order=1) #However, it does not seem to fail for complementary #orientations belonging to [pi/4, 3*pi/4] - approximately # bin_im = rotate(bin_im, 70, order=1) #bin_im = rotate(bin_im, 100, order=1) ##################################################### #Initial data processing fig, ax = plt.subplots(nrows=1, ncols=2) edges = feature.canny(bin_im) data_x, data_y = np.where( edges > 0.5 ) ################################ #Subsample data points for speed indexes = np.arange(0,len(data_x),10) data_x = data_x[indexes] data_y = data_y[indexes] data = np.column_stack([data_x, data_y]) ############################### #Plot original image and border ax[0].imshow(bin_im, cmap=plt.cm.gray) ax[1].imshow(edges, cmap=plt.cm.gray) plt.show() ###################################### #Estimate ellipse parameters from data ellipse = EllipseModel() ellipse.estimate(data) xc, yc, a, b, alpha = ellipse.params ######################################## #Build ellipse from estimated parameters #Rotation matrix R = np.array([ [np.cos(alpha), -np.sin(alpha)], [np.sin(alpha), np.cos(alpha)] ]) a0, a1 = (0, 2*np.pi) angles = np.linspace(a0, a1, 300) X = np.vstack([ np.cos(angles) * a, np.sin(angles) * b]).T ell = np.dot(X, R.T) + (xc, yc) #################### #Plot fitted ellipse fig2, ax2 = plt.subplots(nrows=1, ncols=1) ax2.plot(bin_im.shape[1]-data[:, 1], data[:, 0], '.b', label='Input data') ax2.plot(bin_im.shape[1]-ell[:,1],ell[:,0],'r', label = 'Estimated data') ax2.legend(loc='upper left') # plt.show() #Equation of the lower half of the ellipse should be: #y = y_c + (x-x_c)*tan(alpha) major_axis_length = max(a, b) xx = np.linspace(xc-major_axis_length, xc+major_axis_length, 300) #as an aside question, how should i properly set this range? yy = yc + (xx-xc)*np.tan(alpha) line = np.column_stack([xx, yy]) ax2.plot(bin_im.shape[1]-line[:, 1], line[:, 0], 'g') ax2.set_xlim([0, bin_im.shape[1]]) ax2.set_ylim([0,bin_im.shape[0]]) plt.show() #Build masked semi-ellipse h=bin_im.shape[0] w=bin_im.shape[1] xaxis = np.linspace(0, h-1, h) yaxis = np.linspace(0, w-1, w) yyy, xxx = np.meshgrid(yaxis, xaxis) mask = yyy > yc + (xxx-xc)*np.tan(alpha) new_bin_im = np.zeros_like(bin_im) new_bin_im[mask]=bin_im[mask] fig, ax3 = plt.subplots(nrows=1, ncols=1) ax3.imshow(new_bin_im,cmap=plt.cm.gray) plt.show() -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: bin_mask.png Type: image/png Size: 431 bytes Desc: not available URL: From julien.derr at gmail.com Wed May 13 14:36:37 2015 From: julien.derr at gmail.com (Julien Derr) Date: Wed, 13 May 2015 11:36:37 -0700 (PDT) Subject: problem with io.imread In-Reply-To: <20150504155514.GA2583429@phare.normalesup.org> References: <87383ezhes.fsf@berkeley.edu> <20150504155514.GA2583429@phare.normalesup.org> Message-ID: <995759b3-36d6-404e-acf6-651e827a5e72@googlegroups.com> Hi Emmanuelle, It didn't work, it seemed my installation was completely messy ... I wanted for a long time to move to archlinux, so it was the occasion ... Under arch, the installation was super smooth! and now it works! :-) thanks again for the help, Julien On Monday, May 4, 2015 at 5:55:16 PM UTC+2, Emmanuelle Gouillart wrote: > > did it work? > > On Mon, May 04, 2015 at 08:41:28AM -0700, Julien Derr wrote: > > thanks a lot ! > > > On Sunday, May 3, 2015 at 1:30:39 AM UTC+2, stefanv wrote: > > > Hi Julien > > > On 2015-05-02 05:06:05, Julien Derr wrote: > > > I tried your option; with plugin='matplotlib' and still get the > > > following error message, so it looks like I have indeed a > > > problem with a jpeg decoder, but I checked that I have the > > > package libjpeg8 installed already. > > > Your PIL seems broken. I would completely remove PIL, make sure > > libjpeg and libjpeg-dev is installed, and then install Pillow > > instead. > > > St?fan > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jni.soma at gmail.com Tue May 12 22:20:38 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Wed, 13 May 2015 12:20:38 +1000 Subject: Regionprops Clarification In-Reply-To: References: Message-ID: Indeed, that's what it does, though it's poorly worded in the docs: https://github.com/scikit-image/scikit-image/blob/master/skimage/measure/_regionprops.py#L291 On Tue, May 12, 2015 at 5:26 PM, Joel Frederico wrote: > http://scikit-image.org/docs/dev/api/skimage.measure.html#regionprops > > If we look at the documentation, the attribute weighted_moments_central > uses the region's centroid to subtract when calculating the moments. > Shouldn't it use the region's weighted_centroid values? Does it and it just > doesn't say that? I'm a bit confused. > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From joelfrederico at gmail.com Wed May 13 23:49:01 2015 From: joelfrederico at gmail.com (Joel Frederico) Date: Wed, 13 May 2015 20:49:01 -0700 (PDT) Subject: Regionprops Clarification In-Reply-To: References: Message-ID: <51590caa-3672-4183-bf1f-3d96ed7620d1@googlegroups.com> So it does, thank you! Do you think I / we / you / somebody should report it as a bug? On Tuesday, May 12, 2015 at 7:20:59 PM UTC-7, Juan Nunez-Iglesias wrote: > > Indeed, that's what it does, though it's poorly worded in the docs: > > > https://github.com/scikit-image/scikit-image/blob/master/skimage/measure/_regionprops.py#L291 > > On Tue, May 12, 2015 at 5:26 PM, Joel Frederico > wrote: > >> http://scikit-image.org/docs/dev/api/skimage.measure.html#regionprops >> >> If we look at the documentation, the attribute weighted_moments_central >> uses the region's centroid to subtract when calculating the moments. >> Shouldn't it use the region's weighted_centroid values? Does it and it just >> doesn't say that? I'm a bit confused. >> >> -- >> 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... at googlegroups.com . >> For more options, visit https://groups.google.com/d/optout. >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jni.soma at gmail.com Thu May 14 01:45:37 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Thu, 14 May 2015 15:45:37 +1000 Subject: Regionprops Clarification In-Reply-To: <51590caa-3672-4183-bf1f-3d96ed7620d1@googlegroups.com> References: <51590caa-3672-4183-bf1f-3d96ed7620d1@googlegroups.com> Message-ID: On Thu, May 14, 2015 at 1:49 PM, Joel Frederico wrote: > So it does, thank you! Do you think I / we / you / somebody should report > it as a bug? > Yes. =P If you can raise an issue on GitHub that would be great! A pull request would be even better. If you are not familiar with the process, have a look at our "contributing" documentation ! Thanks again for reporting! Juan. -------------- next part -------------- An HTML attachment was scrubbed... URL: From joelfrederico at gmail.com Thu May 14 20:23:50 2015 From: joelfrederico at gmail.com (Joel Frederico) Date: Thu, 14 May 2015 17:23:50 -0700 (PDT) Subject: Regionprops Clarification In-Reply-To: References: <51590caa-3672-4183-bf1f-3d96ed7620d1@googlegroups.com> Message-ID: <34d9a044-d8ba-4fa6-8b6d-08767dba501b@googlegroups.com> Created a pull request on GitHub: https://github.com/scikit-image/scikit-image/pull/1513 On Wednesday, May 13, 2015 at 10:45:58 PM UTC-7, Juan Nunez-Iglesias wrote: > > On Thu, May 14, 2015 at 1:49 PM, Joel Frederico > wrote: > >> So it does, thank you! Do you think I / we / you / somebody should report >> it as a bug? >> > > Yes. > > =P > > If you can raise an issue on GitHub that would be great! A pull request > would be even better. If you are not familiar with the process, have a look > at our "contributing" documentation > ! > > Thanks again for reporting! > > Juan. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Tue May 19 15:13:30 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Tue, 19 May 2015 12:13:30 -0700 Subject: Fitted Ellipse and its major axis In-Reply-To: <288049b5-06dd-48a7-99d1-a3702b5c33c6@googlegroups.com> References: <288049b5-06dd-48a7-99d1-a3702b5c33c6@googlegroups.com> Message-ID: <877fs4wfc5.fsf@berkeley.edu> Hi Adri?n On 2015-05-12 15:55:45, Adri?n Galdr?n wrote: > I have been trying to fit an ellipse to the border of a roughly > elliptic region. The ellipse fits nicely, but whenever I try to > plot the major axis, it fails. Of note, the approach below works > well for phi \in [pi/4, 3pi/4], but not on other > semi-cuadrants. I think the problem lies on the output of xc, > yc, a, b, alpha = ellipse.params. Here, xc and yc are the > coordinates of the center of the ellipse, a and b are the length > of major and minor axis, and alpha is (I guess) the angle that > the major axis makes against the horizontal axis. Have a look at: http://stackoverflow.com/a/28289147/214686 Perhaps that helps? We also have examples in the documentation on how to plot these axes. Regards St?fan From stefanv at berkeley.edu Tue May 19 15:15:59 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Tue, 19 May 2015 12:15:59 -0700 Subject: Stickers for scipy 2015 Message-ID: <87617owf80.fsf@berkeley.edu> Hi everyone It looks like we may be getting a booth at SciPy2015. I need some ideas for things we can exhibit and show off! Specifically, I am looking for volunteers to: 1. Design and print scikit-image stickers and 2. Design and print a scikit-image poster Please let me know if you are able to help. Thank you in advance! St?fan From daniel.p.bliss at gmail.com Tue May 19 17:00:38 2015 From: daniel.p.bliss at gmail.com (Daniel Bliss) Date: Tue, 19 May 2015 14:00:38 -0700 (PDT) Subject: watershed Message-ID: Hi all, I'm trying to implement the watershed algorithm described in this paper in Python. Looking at the watershed function in skimage.morphology, it's not clear to me whether this function is an exact implementation of Vincent & Soille's method. The docstring for morphology says, "Some ideas taken from . . ." and lists an older (by one year) paper by Soille. "Some ideas" seems to suggest that the implementation is not exact, but where the differences are is not clear. I've read through the morphology source code, as well as the C code it draws upon, but I'm getting tangled up trying to translate between Vincent and Soille's description and the C/Python code in skimage. Can anyone provide a clear and complete explanation of what exactly skimage.morphology.watershed does relative to the Vincent and Soille 1991 method? Many thanks, Dan -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Wed May 20 02:19:05 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Tue, 19 May 2015 23:19:05 -0700 Subject: Stickers for scipy 2015 In-Reply-To: References: <87617owf80.fsf@berkeley.edu> Message-ID: <87egmbvkiu.fsf@berkeley.edu> On 2015-05-19 23:08:10, Juan Nunez-Iglesias wrote: > So cool! > > That's all I have to offer. =P I know there are some artistic talents lurking on the list, secretly keen to help out. We promise not to turn this into a black hole project, folks! All you need: 1. Inkscape 2. A free Saturday afternoon. 3. A desire for scikit-image to be propelled to infinite fame and glory. Reach out, make it happen! St?fan From drake.steven at gmail.com Wed May 20 07:30:32 2015 From: drake.steven at gmail.com (Steven Drake) Date: Wed, 20 May 2015 04:30:32 -0700 (PDT) Subject: shut down the viewer loop? Message-ID: Hello, I've only done a cursory search on this and not finding anything. What terms should be used to search for this? Specifically, here's what I am trying to find out... My script uses viewer.ImageViewer and cycles through several images. The images are closed by clicking on the "X" in the upper right corner. After the last image is closed, the script seems to continue running but does not execute any further commands. My guess is there is a Qt loop still running that must be exited before continuing with the rest of the script. How to stop the loop and continue? Or might there be something else going on? Should I post the script? Steven -------------- next part -------------- An HTML attachment was scrubbed... URL: From vighneshbirodkar at gmail.com Wed May 20 10:50:54 2015 From: vighneshbirodkar at gmail.com (Vighnesh Birodkar) Date: Wed, 20 May 2015 07:50:54 -0700 (PDT) Subject: Stickers for scipy 2015 In-Reply-To: <87egmbvkiu.fsf@berkeley.edu> References: <87617owf80.fsf@berkeley.edu> <87egmbvkiu.fsf@berkeley.edu> Message-ID: <26079b23-c2f3-4ec8-993e-478ca6e1345b@googlegroups.com> I am not much of a creative person, but what exactly are you looking for here ? Pretty looking examples using the API ? On Wednesday, May 20, 2015 at 11:49:10 AM UTC+5:30, stefanv wrote: > > On 2015-05-19 23:08:10, Juan Nunez-Iglesias > > wrote: > > So cool! > > > > That's all I have to offer. =P > > I know there are some artistic talents lurking on the list, > secretly keen to help out. We promise not to turn this into a > black hole project, folks! All you need: > > 1. Inkscape > 2. A free Saturday afternoon. > 3. A desire for scikit-image to be propelled to infinite fame and > glory. > > Reach out, make it happen! > > St?fan > -------------- next part -------------- An HTML attachment was scrubbed... URL: From kevin.keraudren at googlemail.com Wed May 20 12:01:48 2015 From: kevin.keraudren at googlemail.com (Kevin Keraudren) Date: Wed, 20 May 2015 09:01:48 -0700 (PDT) Subject: watershed In-Reply-To: <20150520141346.GA1688446@phare.normalesup.org> References: <20150520141346.GA1688446@phare.normalesup.org> Message-ID: Hi, Regarding the bug you mentioned, I just proposed a fix: https://github.com/scikit-image/scikit-image/issues/803 I found the bug by noticing that the variable "age" was incremented without being used. Of course, as Emmanuelle mentioned, the bug is also in the CellProfiler code, who might be interested in the fix (CC-ed). Kind Regards, Kevin On Wednesday, May 20, 2015 at 3:13:55 PM UTC+1, Emmanuelle wrote: > > Hi Daniel, > > you might want to ask the question on the cell profiler mailing-list as > well since it's their implementation that was included in scikit-image > (cellprofiler-dev >). > > Cheers > Emmanuelle > > On Wed, May 20, 2015 at 04:06:52PM +1000, Juan Nunez-Iglesias wrote: > > Hi Daniel, > > > Actually there's a known bug [1]_ in the scikit-image implementation > (which is > > originally from CellProfiler). I haven't figured it out yet, nor do I > know what > > the exact differences are to the paper. (And most images are > unaffected.) Are > > you looking to use the implementation or are you trying to implement > your own > > as a learning strategy? If the latter, you might want to have a look at > my own > > implementation [2]_ in gala, for which I tried to follow the paper > exactly. But > > it's way slower. > > > Juan. > > > .. [1] https://github.com/scikit-image/scikit-image/issues/803 > > .. [2] > https://github.com/janelia-flyem/gala/blob/master/gala/morpho.py#L212 > > > On Wed, May 20, 2015 at 7:00 AM, Daniel Bliss > wrote: > > > Hi all, > > > I'm trying to implement the watershed algorithm described in this > paper in > > Python. Looking at the watershed function in skimage.morphology, > it's not > > clear to me whether this function is an exact implementation of > Vincent & > > Soille's method. The docstring for morphology says, "Some ideas > taken from > > . . ." and lists an older (by one year) paper by Soille. "Some > ideas" > > seems to suggest that the implementation is not exact, but where the > > differences are is not clear. I've read through the morphology > source > > code, as well as the C code it draws upon, but I'm getting tangled > up > > trying to translate between Vincent and Soille's description and the > C/ > > Python code in skimage. Can anyone provide a clear and complete > > explanation of what exactly skimage.morphology.watershed does > relative to > > the Vincent and Soille 1991 method? > > > Many thanks, > > Dan > -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Wed May 20 18:41:34 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Wed, 20 May 2015 15:41:34 -0700 Subject: Stickers for scipy 2015 In-Reply-To: <26079b23-c2f3-4ec8-993e-478ca6e1345b@googlegroups.com> References: <87617owf80.fsf@berkeley.edu> <87egmbvkiu.fsf@berkeley.edu> <26079b23-c2f3-4ec8-993e-478ca6e1345b@googlegroups.com> Message-ID: <87siaqub1d.fsf@berkeley.edu> Hi Vighnesh On 2015-05-20 07:50:54, Vighnesh Birodkar wrote: > I am not much of a creative person, but what exactly are you > looking for here ? Pretty looking examples using the API ? Yes, if you could help to identify beautiful gallery examples with an impact, that would be useful. Then we put a few of those in an inkscape document, add some text, and voila. St?fan From stefanv at berkeley.edu Wed May 20 19:00:34 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Wed, 20 May 2015 16:00:34 -0700 Subject: shut down the viewer loop? In-Reply-To: References: Message-ID: <87mw0yua5p.fsf@berkeley.edu> On 2015-05-20 04:30:32, Steven Drake wrote: > Should I post the script? That would be helpful, thanks. St?fan From jni.soma at gmail.com Wed May 20 02:06:52 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Wed, 20 May 2015 16:06:52 +1000 Subject: watershed In-Reply-To: References: Message-ID: Hi Daniel, Actually there's a known bug [1]_ in the scikit-image implementation (which is originally from CellProfiler). I haven't figured it out yet, nor do I know what the exact differences are to the paper. (And most images are unaffected.) Are you looking to *use* the implementation or are you trying to implement your own as a learning strategy? If the latter, you might want to have a look at my own implementation [2]_ in gala, for which I tried to follow the paper exactly. But it's way slower. Juan. .. [1] https://github.com/scikit-image/scikit-image/issues/803 .. [2] https://github.com/janelia-flyem/gala/blob/master/gala/morpho.py#L212 On Wed, May 20, 2015 at 7:00 AM, Daniel Bliss wrote: > Hi all, > > I'm trying to implement the watershed algorithm described in this paper > > in Python. Looking at the watershed function in skimage.morphology, it's > not clear to me whether this function is an exact implementation of Vincent > & Soille's method. The docstring for morphology says, "Some ideas taken > from . . ." and lists an older (by one year) paper by Soille. "Some ideas" > seems to suggest that the implementation is not exact, but where the > differences are is not clear. I've read through the morphology source > code, as well as the C code it draws upon, but I'm getting tangled up > trying to translate between Vincent and Soille's description and the > C/Python code in skimage. Can anyone provide a clear and complete > explanation of what exactly skimage.morphology.watershed does relative to > the Vincent and Soille 1991 method? > > Many thanks, > Dan > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jni.soma at gmail.com Wed May 20 02:08:10 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Wed, 20 May 2015 16:08:10 +1000 Subject: Stickers for scipy 2015 In-Reply-To: <87617owf80.fsf@berkeley.edu> References: <87617owf80.fsf@berkeley.edu> Message-ID: So cool! That's all I have to offer. =P Juan. PS: One more thing: the galaxy blob detection example should definitely go on the poster! On Wed, May 20, 2015 at 5:15 AM, Stefan van der Walt wrote: > Hi everyone > > It looks like we may be getting a booth at SciPy2015. I need some ideas > for things we can exhibit and show off! > > Specifically, I am looking for volunteers to: > > 1. Design and print scikit-image stickers and > 2. Design and print a scikit-image poster > > Please let me know if you are able to help. > > Thank you in advance! > > St?fan > > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From emmanuelle.gouillart at nsup.org Wed May 20 10:13:46 2015 From: emmanuelle.gouillart at nsup.org (Emmanuelle Gouillart) Date: Wed, 20 May 2015 16:13:46 +0200 Subject: watershed In-Reply-To: References: Message-ID: <20150520141346.GA1688446@phare.normalesup.org> Hi Daniel, you might want to ask the question on the cell profiler mailing-list as well since it's their implementation that was included in scikit-image (cellprofiler-dev ). Cheers Emmanuelle On Wed, May 20, 2015 at 04:06:52PM +1000, Juan Nunez-Iglesias wrote: > Hi Daniel, > Actually there's a known bug [1]_ in the scikit-image implementation (which is > originally from CellProfiler). I haven't figured it out yet, nor do I know what > the exact differences are to the paper. (And most images are unaffected.) Are > you looking to use the implementation or are you trying to implement your own > as a learning strategy? If the latter, you might want to have a look at my own > implementation [2]_ in gala, for which I tried to follow the paper exactly. But > it's way slower. > Juan. > .. [1]??https://github.com/scikit-image/scikit-image/issues/803 > .. [2]??https://github.com/janelia-flyem/gala/blob/master/gala/morpho.py#L212 > On Wed, May 20, 2015 at 7:00 AM, Daniel Bliss wrote: > Hi all, > I'm trying to implement the watershed algorithm described in this paper in > Python.?? Looking at the watershed function in skimage.morphology, it's not > clear to me whether this function is an exact implementation of Vincent & > Soille's method.?? The docstring for morphology says, "Some ideas taken from > . . ." and lists an older (by one year) paper by Soille. ??"Some ideas" > seems to suggest that the implementation is not exact, but where the > differences are is not clear.?? I've read through the morphology source > code, as well as the C code it draws upon, but I'm getting tangled up > trying to translate between Vincent and Soille's description and the C/ > Python code in skimage.?? Can anyone provide a clear and complete > explanation of what exactly skimage.morphology.watershed does relative to > the Vincent and Soille 1991 method? > Many thanks, > Dan From kevin.keraudren at googlemail.com Wed May 20 11:57:39 2015 From: kevin.keraudren at googlemail.com (Kevin Keraudren) Date: Wed, 20 May 2015 16:57:39 +0100 Subject: watershed In-Reply-To: <20150520141346.GA1688446@phare.normalesup.org> References: <20150520141346.GA1688446@phare.normalesup.org> Message-ID: <555CAEF3.7080604@googlemail.com> Hi, Regarding the bug you mentioned, I just proposed a fix: https://github.com/scikit-image/scikit-image/issues/803 I found the bug by noticing that the variable "age" was incremented without being used. Of course, as Emmanuelle mentioned, the bug is also in the CellProfiler code, who might be interested in the fix (CC-ed). Kind Regards, Kevin Le 20/05/2015 15:13, Emmanuelle Gouillart a ?crit : > Hi Daniel, > > you might want to ask the question on the cell profiler mailing-list as > well since it's their implementation that was included in scikit-image > (cellprofiler-dev ). > > Cheers > Emmanuelle > > On Wed, May 20, 2015 at 04:06:52PM +1000, Juan Nunez-Iglesias wrote: >> Hi Daniel, >> Actually there's a known bug [1]_ in the scikit-image implementation (which is >> originally from CellProfiler). I haven't figured it out yet, nor do I know what >> the exact differences are to the paper. (And most images are unaffected.) Are >> you looking to use the implementation or are you trying to implement your own >> as a learning strategy? If the latter, you might want to have a look at my own >> implementation [2]_ in gala, for which I tried to follow the paper exactly. But >> it's way slower. >> Juan. >> .. [1] https://github.com/scikit-image/scikit-image/issues/803 >> .. [2] https://github.com/janelia-flyem/gala/blob/master/gala/morpho.py#L212 >> On Wed, May 20, 2015 at 7:00 AM, Daniel Bliss wrote: >> Hi all, >> I'm trying to implement the watershed algorithm described in this paper in >> Python. Looking at the watershed function in skimage.morphology, it's not >> clear to me whether this function is an exact implementation of Vincent & >> Soille's method. The docstring for morphology says, "Some ideas taken from >> . . ." and lists an older (by one year) paper by Soille. "Some ideas" >> seems to suggest that the implementation is not exact, but where the >> differences are is not clear. I've read through the morphology source >> code, as well as the C code it draws upon, but I'm getting tangled up >> trying to translate between Vincent and Soille's description and the C/ >> Python code in skimage. Can anyone provide a clear and complete >> explanation of what exactly skimage.morphology.watershed does relative to >> the Vincent and Soille 1991 method? >> Many thanks, >> Dan From drake.steven at gmail.com Thu May 21 06:58:14 2015 From: drake.steven at gmail.com (Steven Drake) Date: Thu, 21 May 2015 03:58:14 -0700 (PDT) Subject: shut down the viewer loop? In-Reply-To: <87mw0yua5p.fsf@berkeley.edu> References: <87mw0yua5p.fsf@berkeley.edu> Message-ID: <97e751d2-b67d-4ff3-aa08-d86eccc50268@googlegroups.com> OK, this is a trimmed down version This is the part which cycles through a list of images and displays them on at a time. Each displayed image is closed by clicking the "X" on the frame: for item in existing_data_list: jpeg_name = item['jpeg'] jpeg_set = item['set_type'] jpeg_path = "" if jpeg_set == "TRAIN": jpeg_path = train_data_source_path elif jpeg_set == "TEST": jpeg_path = test_data_source_path else: break jpeg_fullname = os.path.join(jpeg_path, jpeg_name) jpeg_image = ndimage.imread(jpeg_fullname, mode="RGB") an_image = ski_viewer.ImageViewer(jpeg_image) an_image.show() here's the list it cycles through: [{'set_type': 'TRAIN', 'predicted': 9999, 'jpeg': '10_left.jpeg', 'correct': 'unknown', 'level': '0'}, {'set_type': 'TRAIN', 'predicted': 9999, 'jpeg': '10_right.jpeg', 'correct': 'unknown', 'level': '0'}, {'set_type': 'TEST', 'predicted': 9999, 'jpeg': '11_left.jpeg', 'correct': 'unknown', 'level': 9999}, {'set_type': 'TEST', 'predicted': 9999, 'jpeg': '11_right.jpeg', 'correct': 'unknown', 'level': 9999}, {'set_type': 'TEST', 'predicted': 9999, 'jpeg': '12_left.jpeg', 'correct': 'unknown', 'level': 9999}, {'set_type': 'TEST', 'predicted': 9999, 'jpeg': '12_right.jpeg', 'correct': 'unknown', 'level': 9999}] Attached a zip file with the jpegs. You'll need to change the script variables train_data_source_path and test_data_source_path to suit your pathing. Steven -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: question_data.7z Type: application/x-7z-compressed Size: 5980427 bytes Desc: not available URL: From stefanv at berkeley.edu Fri May 22 00:19:04 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Thu, 21 May 2015 21:19:04 -0700 Subject: shut down the viewer loop? In-Reply-To: <97e751d2-b67d-4ff3-aa08-d86eccc50268@googlegroups.com> References: <87mw0yua5p.fsf@berkeley.edu> <97e751d2-b67d-4ff3-aa08-d86eccc50268@googlegroups.com> Message-ID: <87382p1byf.fsf@berkeley.edu> Hi Steven On 2015-05-21 03:58:14, Steven Drake wrote: > OK, this is a trimmed down version Attached, find a fully functional code snippet. This exits on my machine as soon as it is done (but I didn't modify your code at all). Regards St?fan -------------- next part -------------- A non-text attachment was scrubbed... Name: test.py Type: text/x-python Size: 1205 bytes Desc: not available URL: From drake.steven at gmail.com Fri May 22 07:00:46 2015 From: drake.steven at gmail.com (Steven Drake) Date: Fri, 22 May 2015 04:00:46 -0700 (PDT) Subject: shut down the viewer loop? In-Reply-To: <87382p1byf.fsf@berkeley.edu> References: <87mw0yua5p.fsf@berkeley.edu> <97e751d2-b67d-4ff3-aa08-d86eccc50268@googlegroups.com> <87382p1byf.fsf@berkeley.edu> Message-ID: <27d9a1b2-e8dd-4ca9-847a-ff88d05e494d@googlegroups.com> Stefan, I started digging deeper and it appears some of the jpegs are the problem. For unknown reason they are causing viewer.ImageViewer to hang. Thank You. I would still be spinning my wheels searching in the wrong place. Steven On Friday, May 22, 2015 at 12:19:08 AM UTC-4, stefanv wrote: > Hi Steven > > On 2015-05-21 03:58:14, Steven Drake > > wrote: > > OK, this is a trimmed down version > > Attached, find a fully functional code snippet. This exits on my > machine as soon as it is done (but I didn't modify your code at > all). > > Regards > St?fan > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From drake.steven at gmail.com Fri May 22 07:31:04 2015 From: drake.steven at gmail.com (Steven Drake) Date: Fri, 22 May 2015 04:31:04 -0700 (PDT) Subject: shut down the viewer loop? In-Reply-To: <87382p1byf.fsf@berkeley.edu> References: <87mw0yua5p.fsf@berkeley.edu> <97e751d2-b67d-4ff3-aa08-d86eccc50268@googlegroups.com> <87382p1byf.fsf@berkeley.edu> Message-ID: Stefan, As a follow-on to my last post... it was not the jpegs directly causing the problem. After about 18 images displayed, viewer.ImageViewer started hanging. I added into the script a line to delete the ImageViewer object after closing the window (clicking "X") and the script kept on going to finish. Seems to me like the ImageViewer objects do not completely go away after closing and at some point a memory limit -- or number of allowed ImageViewer objects -- is reached. Any thoughts? Steven -------------- next part -------------- An HTML attachment was scrubbed... URL: From linux.experi at gmail.com Fri May 22 14:19:54 2015 From: linux.experi at gmail.com (user783746) Date: Fri, 22 May 2015 11:19:54 -0700 (PDT) Subject: Advice for Audio classifier based on Voice Activity Detection Message-ID: I am writting a program to classify recorded audio phone calls files (wav) which contain atleast some Human Voice or Non Voice (only DTMF, Dialtones, ringtones, noise). I tried implementing simple VAD (voice activity detector) using ZCR (zero crossing rate) & calculating Energy, but these parameters confuse with DTMF, Dialtones files with Voice. I also tried implementing a machine learning based approach using SVM (Support Vector Machine) and MFCC coefficients. The results were worse than previous approach. I need someone to advice me little on this domain, I have no previous experience in machine learning or AI. I am willing to put in good amount of time in this domain. I am comfortable working in MATLAB, scipy, numpy, scikit-learn, python. Thank you -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Fri May 22 15:04:26 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Fri, 22 May 2015 12:04:26 -0700 Subject: Advice for Audio classifier based on Voice Activity Detection In-Reply-To: References: Message-ID: <87siaozb5x.fsf@berkeley.edu> On 2015-05-22 11:19:54, user783746 wrote: > I am writting a program to classify recorded audio phone calls > files (wav) which contain atleast some Human Voice or Non Voice > (only DTMF, Dialtones, ringtones, noise). I tried implementing > simple VAD (voice activity detector) using ZCR (zero crossing > rate) & calculating Energy, but these parameters confuse with > DTMF, Dialtones files with Voice. > > I also tried implementing a machine learning based approach > using SVM (Support Vector Machine) and MFCC coefficients. The > results were worse than previous approach. The problem description is pretty vague, but I guess that you're better off asking on the scipy or scikit-learn lists. Regards St?fan From silvertrumpet999 at gmail.com Fri May 22 15:12:26 2015 From: silvertrumpet999 at gmail.com (Josh Warner) Date: Fri, 22 May 2015 12:12:26 -0700 (PDT) Subject: Advice for Audio classifier based on Voice Activity Detection In-Reply-To: References: Message-ID: <7552e36a-85ce-4ea6-b54e-042baca81eb0@googlegroups.com> Your problem needs time-frequency analysis . Generate some waterfall plots of time vs. frequency using windowed Fourier transforms. Inspect those, and/or use them them as input into your learning approaches. Depending on the approach you want to use, the waterfall plots can be analyzed like images with some caveats as the rows and columns represent entirely different physical measurements. That's the main place where scikit-image could potentially assist. This is its own sub-field of digital signal processing; now that you know the keyword to search against you can peruse a large body of literature to assist with your project. Josh On Friday, May 22, 2015 at 2:00:58 PM UTC-5, user783746 wrote: > > I am writting a program to classify recorded audio phone calls files (wav) > which contain atleast some Human Voice or Non Voice (only DTMF, Dialtones, > ringtones, noise). I tried implementing simple VAD (voice activity > detector) using ZCR (zero crossing rate) & calculating Energy, but these > parameters confuse with DTMF, Dialtones files with Voice. > > I also tried implementing a machine learning based approach using SVM > (Support Vector Machine) and MFCC coefficients. The results were worse than > previous approach. > > I need someone to advice me little on this domain, I have no previous > experience in machine learning or AI. I am willing to put in good amount of > time in this domain. > > I am comfortable working in MATLAB, scipy, numpy, scikit-learn, python. > > Thank you > -------------- next part -------------- An HTML attachment was scrubbed... URL: From linux.experi at gmail.com Fri May 22 15:23:31 2015 From: linux.experi at gmail.com (user783746) Date: Fri, 22 May 2015 12:23:31 -0700 (PDT) Subject: Advice for Audio classifier based on Voice Activity Detection In-Reply-To: References: Message-ID: Thank you Josh On Saturday, May 23, 2015 at 12:30:58 AM UTC+5:30, user783746 wrote: > > I am writting a program to classify recorded audio phone calls files (wav) > which contain atleast some Human Voice or Non Voice (only DTMF, Dialtones, > ringtones, noise). I tried implementing simple VAD (voice activity > detector) using ZCR (zero crossing rate) & calculating Energy, but these > parameters confuse with DTMF, Dialtones files with Voice. > > I also tried implementing a machine learning based approach using SVM > (Support Vector Machine) and MFCC coefficients. The results were worse than > previous approach. > > I need someone to advice me little on this domain, I have no previous > experience in machine learning or AI. I am willing to put in good amount of > time in this domain. > > I am comfortable working in MATLAB, scipy, numpy, scikit-learn, python. > > Thank you > -------------- next part -------------- An HTML attachment was scrubbed... URL: From steven.silvester at gmail.com Fri May 22 20:51:16 2015 From: steven.silvester at gmail.com (Steven Silvester) Date: Fri, 22 May 2015 17:51:16 -0700 (PDT) Subject: shut down the viewer loop? In-Reply-To: References: Message-ID: Steven, Can you try calling show() only on the *last* ImageViewer you create? On Wednesday, May 20, 2015 at 6:41:12 AM UTC-5, Steven Drake wrote: > > Hello, > > I've only done a cursory search on this and not finding anything. What > terms should be used to search for this? > > Specifically, here's what I am trying to find out... > > My script uses viewer.ImageViewer and cycles through several images. The > images are closed by clicking on the "X" in the upper right corner. After > the last image is closed, the script seems to continue running but does not > execute any further commands. > > My guess is there is a Qt loop still running that must be exited before > continuing with the rest of the script. How to stop the loop and continue? > Or might there be something else going on? > > Should I post the script? > > Steven > -------------- next part -------------- An HTML attachment was scrubbed... URL: From warmspringwinds at gmail.com Sat May 23 05:24:20 2015 From: warmspringwinds at gmail.com (Daniil Pakhomov) Date: Sat, 23 May 2015 02:24:20 -0700 (PDT) Subject: GSoC 2015 - Accepted Students In-Reply-To: References: Message-ID: <2907a03d-6783-491e-9590-9d65ea4a5096@googlegroups.com> Hello, @Vignesh and @Stefan. This is my introduction blog post: http://warmspringwinds.github.io/gsoc/face_detection/scikit_image/2015/05/22/google-summer-of-code-patent-free-face-detection-for-scikit-image-in-python.-introduction/ Please, have a look at it. This is the respective repository where you can make suggestions on the changes for my blog post: https://github.com/warmspringwinds/warmspringwinds.github.io Thank you. ???????, 1 ??? 2015 ?., 21:17:19 UTC+2 ???????????? Vighnesh Birodkar ???????: > > Hello Everyone > > Please join in in congratulating Aman and Daniil for being a part of GSoC > 2015 for scikit-image. We look forward to an awesome summer. > > I'd like to request Aman and Daniil to join the GSoC specific mailing > list[1]. This will mainly hold project specific discussions and updates. > Anyone interested in the projects can also join and share their thoughts. > > As part of PSF, GSoC students are required to write a blog post every 2 > weeks. It would be nice if we could have an introductory blog post about > each project during the community bonding period. About Daniil's project > specifically, I'd like to see an overview of his face detection algorithm. > Specifically, I'd like to see a pictorial representation of MB-LBP and what > exactly the Viola-Jones method has patented. > > Please keep in mind, that these blog posts will not only be read by the > mentors, but the general public as well, so they should be easily > understandable and also include all the necessary details. A well > documented face detection implementation would help gain a lot of traction > for the project. Similarly (although I'm not clear about the specifics) if > the design and strategy of the cython implementation of ndimage is well > documented and explained, it will help the future maintainers enormously. > > For all your blog posts, keep this scenario in mind : A user googles "Face > Detection / Python image filters ", and sequentially clicks each link till > she/he finds what he wants. > > I'd also request Stefan and Ralf to chip in with their thoughts and share > if they agree/disagree with my ideas about these blog posts (I'm trying to > get a hang of this mentoring thing). Also, Juan can share his thoughts, for > he use to always push me (and rightfully so) towards writing better last > year. > > As a reference, see the blog feeds from last year written by Rishabh and > me. You can also dig into previous years' GSoC archives to find more. > > Thanks > Vighnesh > > [1] : https://groups.google.com/forum/#!forum/skimage-gsoc > [2] : https://sharkysoc14.wordpress.com/ > [3] : https://vcansimplify.wordpress.com/tag/gsoc/ > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From kwiechen1 at gmail.com Mon May 25 14:49:52 2015 From: kwiechen1 at gmail.com (Kai Wiechen) Date: Mon, 25 May 2015 11:49:52 -0700 (PDT) Subject: missing kernel or block size in equalize_adapthist Message-ID: <4b57c4b4-1f5f-4cf5-a4e6-3b14e657da2f@googlegroups.com> Hello, I am using equalize_adapthist after color deconvolution and some morphological operations to enhance local contrast of nuclei in histological images. There are different results when applying equalize_adapthist when using small patches and larger size whole images (left small patch, right large size whole image). The CLAHE implementation in ImageJ has a 'block size' parameter and the implementation in Pixinsight has a 'Kernel size' parameter to get results depending on the local context. Is it possible to 'simulate' this behaviour with scikit-image? Best regards, Kai -------------- next part -------------- An HTML attachment was scrubbed... URL: From jaime.frio at gmail.com Mon May 25 15:37:29 2015 From: jaime.frio at gmail.com (=?UTF-8?Q?Jaime_Fern=C3=A1ndez_del_R=C3=ADo?=) Date: Mon, 25 May 2015 12:37:29 -0700 Subject: Repository for GSoC project Message-ID: Hi all, This year I am mentoring Aman for one of the GSoC projects we have underway, "Rewriting ndimage in Cython." By its very nature it doesn't conform very well to the "many small pull requests" model: from the point of view of scipy, things are going to be broken up until almost the very last commit. I am not sure what the best way to set up a collaborative code development environment would be, and so am asking for the collective wisdom to help guide us. Aman could simply create one ginormous pull request that will grow, and grow, and not be merged until everything was ready. I don't like this idea too much, as it is going to eventually be a confusing mess, and I think it would also make it difficult for others than Aman (that would mostly be me) to contribute code. I think we could also use a branch, either on my fork of scipy or on Aman's, as the repository on which development would happen, and against which PRs would be created, and once completed send a single PR to the main scipy repo. This may work, but I don't like it much either. What probably makes more sense is to create a new branch **in the main scipy repository**, and have PRs sent and merged against that branch, which would eventually be merged with master upon completion. NumPy seems to have a couple such experimental branches ('with_maskna' and 'enable_separate_by_default'), although there is none in SciPy that I see. This would also allow us to keep the project in a controlled environment, even if by the end of the summer not every single bit of ndimage has been ported. If this third path is really the preferred way of doing things, I could probably set things up myself (Ralf gave me commit rights when I became a mentor for this project), but I'd like to hear what others think, before abusing my powers. Thanks! Jaime -- (\__/) ( O.o) ( > <) Este es Conejo. Copia a Conejo en tu firma y ay?dale en sus planes de dominaci?n mundial. -------------- next part -------------- An HTML attachment was scrubbed... URL: From steven.silvester at gmail.com Mon May 25 18:54:34 2015 From: steven.silvester at gmail.com (Steven Silvester) Date: Mon, 25 May 2015 15:54:34 -0700 (PDT) Subject: missing kernel or block size in equalize_adapthist In-Reply-To: <4b57c4b4-1f5f-4cf5-a4e6-3b14e657da2f@googlegroups.com> References: <4b57c4b4-1f5f-4cf5-a4e6-3b14e657da2f@googlegroups.com> Message-ID: <903b3345-0f2d-4c5f-85bf-2c22fa4942a9@googlegroups.com> Hi Kai, Do the `n_tiles*` arguments not meet your needs? By defining the number of tiles in X and Y, you are equivalently setting a block size. http://scikit-image.org/docs/stable/api/skimage.exposure.html#equalize-adapthist Regards, Steve On Monday, May 25, 2015 at 1:49:53 PM UTC-5, Kai Wiechen wrote: > > > > Hello, > > I am using equalize_adapthist after color deconvolution and some > morphological operations to enhance local contrast of nuclei in > histological images. There are different results when applying > equalize_adapthist when using small patches and larger size whole images > (left small patch, right large size whole image). The CLAHE implementation > in ImageJ has a 'block size' parameter and the implementation in Pixinsight > has a 'Kernel size' parameter to get results depending on the local > context. Is it possible to 'simulate' this behaviour with scikit-image? > > Best regards, > > Kai > > > > > > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From jni.soma at gmail.com Tue May 26 00:08:10 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Mon, 25 May 2015 21:08:10 -0700 (PDT) Subject: Repository for GSoC project In-Reply-To: References: Message-ID: <1432613289940.37093e0a@Nodemailer> Hey Jaime, (et al) This still leaves open the question of *how* you will merge things in this magical branch...? What will constitute a sufficiently-complete PR against this branch that would not be so in master? (I do 100% agree that a ginormous PR is not the way to go, though!) Juan. On Tue, May 26, 2015 at 5:37 AM, Jaime Fern?ndez del R?o wrote: > Hi all, > This year I am mentoring Aman for one of the GSoC projects we have > underway, "Rewriting ndimage in Cython." By its very nature it doesn't > conform very well to the "many small pull requests" model: from the point > of view of scipy, things are going to be broken up until almost the very > last commit. I am not sure what the best way to set up a collaborative code > development environment would be, and so am asking for the collective > wisdom to help guide us. > Aman could simply create one ginormous pull request that will grow, and > grow, and not be merged until everything was ready. I don't like this idea > too much, as it is going to eventually be a confusing mess, and I think it > would also make it difficult for others than Aman (that would mostly be me) > to contribute code. > I think we could also use a branch, either on my fork of scipy or on > Aman's, as the repository on which development would happen, and against > which PRs would be created, and once completed send a single PR to the main > scipy repo. This may work, but I don't like it much either. > What probably makes more sense is to create a new branch **in the main > scipy repository**, and have PRs sent and merged against that branch, which > would eventually be merged with master upon completion. NumPy seems to have > a couple such experimental branches ('with_maskna' and > 'enable_separate_by_default'), although there is none in SciPy that I see. > This would also allow us to keep the project in a controlled environment, > even if by the end of the summer not every single bit of ndimage has been > ported. > If this third path is really the preferred way of doing things, I could > probably set things up myself (Ralf gave me commit rights when I became a > mentor for this project), but I'd like to hear what others think, before > abusing my powers. > Thanks! > Jaime > -- > (\__/) > ( O.o) > ( > <) Este es Conejo. Copia a Conejo en tu firma y ay?dale en sus planes > de dominaci?n mundial. > -- > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. -------------- next part -------------- An HTML attachment was scrubbed... URL: From kwiechen1 at gmail.com Tue May 26 02:58:55 2015 From: kwiechen1 at gmail.com (Kai Wiechen) Date: Mon, 25 May 2015 23:58:55 -0700 (PDT) Subject: missing kernel or block size in equalize_adapthist In-Reply-To: <903b3345-0f2d-4c5f-85bf-2c22fa4942a9@googlegroups.com> References: <4b57c4b4-1f5f-4cf5-a4e6-3b14e657da2f@googlegroups.com> <903b3345-0f2d-4c5f-85bf-2c22fa4942a9@googlegroups.com> Message-ID: Hi Steve, but these parameters are limited to the range 1..16. The small test patches are 150x150 pixels, the complete images have 1920x1448 pixels. I have applied equalize_adapthist here with default parameters (n_tiles=8). Best regards, Kai On Tuesday, May 26, 2015 at 12:54:34 AM UTC+2, Steven Silvester wrote: > > Hi Kai, > > Do the `n_tiles*` arguments not meet your needs? By defining the number > of tiles in X and Y, you are equivalently setting a block size. > > > http://scikit-image.org/docs/stable/api/skimage.exposure.html#equalize-adapthist > > > Regards, > > Steve > > On Monday, May 25, 2015 at 1:49:53 PM UTC-5, Kai Wiechen wrote: >> >> >> >> Hello, >> >> I am using equalize_adapthist after color deconvolution and some >> morphological operations to enhance local contrast of nuclei in >> histological images. There are different results when applying >> equalize_adapthist when using small patches and larger size whole images >> (left small patch, right large size whole image). The CLAHE implementation >> in ImageJ has a 'block size' parameter and the implementation in Pixinsight >> has a 'Kernel size' parameter to get results depending on the local >> context. Is it possible to 'simulate' this behaviour with scikit-image? >> >> Best regards, >> >> Kai >> >> >> >> >> >> >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From vighneshbirodkar at gmail.com Tue May 26 05:17:56 2015 From: vighneshbirodkar at gmail.com (Vighnesh Birodkar) Date: Tue, 26 May 2015 02:17:56 -0700 (PDT) Subject: Repository for GSoC project In-Reply-To: References: Message-ID: <78b496db-c70c-4a16-97a8-fa9115716280@googlegroups.com> Hello Correct me if I am wrong here, but I remember a discussion where someone said ndimage module is divided into ~6 files ? How about we rewrite one file at a time ? Also, I'd like Jaime to look at this : https://github.com/scikit-image/scikit-image/pull/1510 Does this partly overlap with Aman's project ? Thanks Vighnesh On Tuesday, May 26, 2015 at 1:07:31 AM UTC+5:30, Jaime Fern?ndez del R?o wrote: > > Hi all, > > This year I am mentoring Aman for one of the GSoC projects we have > underway, "Rewriting ndimage in Cython." By its very nature it doesn't > conform very well to the "many small pull requests" model: from the point > of view of scipy, things are going to be broken up until almost the very > last commit. I am not sure what the best way to set up a collaborative code > development environment would be, and so am asking for the collective > wisdom to help guide us. > > Aman could simply create one ginormous pull request that will grow, and > grow, and not be merged until everything was ready. I don't like this idea > too much, as it is going to eventually be a confusing mess, and I think it > would also make it difficult for others than Aman (that would mostly be me) > to contribute code. > > I think we could also use a branch, either on my fork of scipy or on > Aman's, as the repository on which development would happen, and against > which PRs would be created, and once completed send a single PR to the main > scipy repo. This may work, but I don't like it much either. > > What probably makes more sense is to create a new branch **in the main > scipy repository**, and have PRs sent and merged against that branch, which > would eventually be merged with master upon completion. NumPy seems to have > a couple such experimental branches ('with_maskna' and > 'enable_separate_by_default'), although there is none in SciPy that I see. > This would also allow us to keep the project in a controlled environment, > even if by the end of the summer not every single bit of ndimage has been > ported. > > If this third path is really the preferred way of doing things, I could > probably set things up myself (Ralf gave me commit rights when I became a > mentor for this project), but I'd like to hear what others think, before > abusing my powers. > > Thanks! > > Jaime > > -- > (\__/) > ( O.o) > ( > <) Este es Conejo. Copia a Conejo en tu firma y ay?dale en sus planes > de dominaci?n mundial. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From edwardando at gmail.com Tue May 26 08:41:50 2015 From: edwardando at gmail.com (Edward Ando) Date: Tue, 26 May 2015 05:41:50 -0700 (PDT) Subject: PhD Thesis Proposal Message-ID: Dear All, I'm writing to you with a piece of information that may interest you: Together with some colleagues we have secured funding for a PhD in image analysis applied to Geomaterials in Grenoble (France), starting in October 2015. More information, and the addresses to write to for the application here: https://docs.google.com/document/d/1mGzZT7GRAhdtk1_Y2HwhfVkzU_IvHhyOZVJYkVAPJMo Thanks for your attention, Edward And? -------------- next part -------------- An HTML attachment was scrubbed... URL: From jaime.frio at gmail.com Tue May 26 12:00:12 2015 From: jaime.frio at gmail.com (=?UTF-8?Q?Jaime_Fern=C3=A1ndez_del_R=C3=ADo?=) Date: Tue, 26 May 2015 09:00:12 -0700 Subject: Repository for GSoC project In-Reply-To: <78b496db-c70c-4a16-97a8-fa9115716280@googlegroups.com> References: <78b496db-c70c-4a16-97a8-fa9115716280@googlegroups.com> Message-ID: Thanks everyone for the feedback. After some more thinking, I have asked Aman to start sending PRs directly against the current master of scipy. To differentiate new from old implementations, we'll probably just prepend all file names with "ndi_" instead of the current "ni_". We will eventually start removing old functionality that has been replaced, but for quite sometime they will both coexist. This means that at some point in time, e.g. calling `ndimage.filters.uniform_filter1d` may use the new Cython code paths, while `ndimage.filters.minimum_filter1d` may still be using the old C code paths. We will not merge anything, answering Juan's question, until we have tests that validate that functionality. That does not mean that the functionality will be available to end users just yet, e.g. we may merge an iterator before any top level function that uses it. But if it does not have tests, it does not get merged. If anyone has a very strong opinion against this approach, please speak up now! Jaime -- (\__/) ( O.o) ( > <) Este es Conejo. Copia a Conejo en tu firma y ay?dale en sus planes de dominaci?n mundial. -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Tue May 26 20:03:49 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Tue, 26 May 2015 17:03:49 -0700 Subject: PhD Thesis Proposal In-Reply-To: References: Message-ID: <877fruzy1m.fsf@berkeley.edu> Hi Edward On 2015-05-26 05:41:50, Edward Ando wrote: > I'm writing to you with a piece of information that may interest > you: Together with some colleagues we have secured funding for > a PhD in image analysis applied to Geomaterials in Grenoble > (France), starting in October 2015. Out of curiosity, is scikit-image a potential platform for this project? If so, great! If not, the reason would be interesting to know. Regards St?fan From steven.silvester at gmail.com Tue May 26 21:16:57 2015 From: steven.silvester at gmail.com (Steven Silvester) Date: Tue, 26 May 2015 20:16:57 -0500 Subject: missing kernel or block size in equalize_adapthist In-Reply-To: References: <4b57c4b4-1f5f-4cf5-a4e6-3b14e657da2f@googlegroups.com> <903b3345-0f2d-4c5f-85bf-2c22fa4942a9@googlegroups.com> Message-ID: Hi Kai, Is the issue that you want finer grain control of the block size? If so, that would require a substantial rewrite. Please feel free to open an issue on Github if that is the case. Regards, Steve > On May 26, 2015, at 1:58 AM, Kai Wiechen wrote: > > Hi Steve, > > but these parameters are limited to the range 1..16. The small test patches are 150x150 pixels, the complete images have 1920x1448 pixels. I have applied equalize_adapthist here with default parameters (n_tiles=8). > > Best regards, > > Kai > > On Tuesday, May 26, 2015 at 12:54:34 AM UTC+2, Steven Silvester wrote: > Hi Kai, > > Do the `n_tiles*` arguments not meet your needs? By defining the number of tiles in X and Y, you are equivalently setting a block size. > > http://scikit-image.org/docs/stable/api/skimage.exposure.html#equalize-adapthist > > > Regards, > > Steve > > On Monday, May 25, 2015 at 1:49:53 PM UTC-5, Kai Wiechen wrote: > Hello, > > I am using equalize_adapthist after color deconvolution and some morphological operations to enhance local contrast of nuclei in histological images. There are different results when applying equalize_adapthist when using small patches and larger size whole images (left small patch, right large size whole image). The CLAHE implementation in ImageJ has a 'block size' parameter and the implementation in Pixinsight has a 'Kernel size' parameter to get results depending on the local context. Is it possible to 'simulate' this behaviour with scikit-image? > > Best regards, > > Kai > > > > > > > > > -- > 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/2elntOClGOY/unsubscribe . > To unsubscribe from this group and all its topics, send an email to scikit-image+unsubscribe at googlegroups.com . > For more options, visit https://groups.google.com/d/optout . -------------- next part -------------- An HTML attachment was scrubbed... URL: From jaime.frio at gmail.com Wed May 27 13:57:37 2015 From: jaime.frio at gmail.com (=?UTF-8?Q?Jaime_Fern=C3=A1ndez_del_R=C3=ADo?=) Date: Wed, 27 May 2015 10:57:37 -0700 Subject: [ndimage-gsoc2015] Weekly Summary 2015/05/25 Message-ID: Hi all, Aman and I had our first weekly sync meeting yesterday. After each of these weekly calls, I have asked Aman to publish a short summary, just like this one, on progress to date and plans for the week. If you would like to participate on one of these weekly meetings, just let us know. *Design Choices* 1. Use the NumPy iterator (numpy.flatiter in Python or PyArrayIterObject in C) for all basic array iteration, rather than a custom implementation as ndimage currently does. *Plan for the week* 1. Learn to use the NumPy iterator in Cython properly. 2. Send a first pull request against ndimage that: 1. Provides supporting Cython code for basic array iteration. 2. Translates an ndimage function that only requires basic iteration, to Cython. This will probably be find_objects from the measurements submodule. 3. Document the progress in a blog post. *Deliverables from last week* 1. None yet. Best, Jaime -- (\__/) ( O.o) ( > <) Este es Conejo. Copia a Conejo en tu firma y ay?dale en sus planes de dominaci?n mundial. -------------- next part -------------- An HTML attachment was scrubbed... URL: From as.zsuzsanna at gmail.com Thu May 28 01:30:42 2015 From: as.zsuzsanna at gmail.com (=?UTF-8?Q?Zsuzsanna_P=C3=BCsp=C3=B6ki?=) Date: Wed, 27 May 2015 22:30:42 -0700 (PDT) Subject: skimage.feature Message-ID: <77ff627c-7334-4f28-af91-d98d9a0d12e5@googlegroups.com> I am trying to use the blob_dog, blob_log, blob_doh from skimage.feature on some synthetic data. 1. Why do you obtain such different results for the LoG and DoG on your sample data: http://scikit-image.org/docs/dev/auto_examples/plot_blob.html ? I would expect something very similar. 2. I am running the following line on the attached image: blobs_log = blob_log(image_gray, overlap = 0, min_sigma=3, max_sigma=25, num_sigma=100, threshold=.1) I cannot arrive to detect precisely the size of the bigger spots (though the range I put is enough, hence the huge spot). The increase in num_sigma does not solve the issue. I would expect perfect results here. 3. Is there an easy way to visualize the N best detections? -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Screen Shot 2015-05-27 at 14.36.47.png Type: image/png Size: 57818 bytes Desc: not available URL: From kwiechen1 at gmail.com Thu May 28 05:15:09 2015 From: kwiechen1 at gmail.com (Kai Wiechen) Date: Thu, 28 May 2015 02:15:09 -0700 (PDT) Subject: skimage.feature In-Reply-To: <77ff627c-7334-4f28-af91-d98d9a0d12e5@googlegroups.com> References: <77ff627c-7334-4f28-af91-d98d9a0d12e5@googlegroups.com> Message-ID: <3f2ef3f6-cfed-4fb9-9646-d6d7ded80388@googlegroups.com> Have you tried to modify the overlap or the max_sigma parameters? There is an internal function _prune_blobs to remove overlapping signals favoring larger ones. Regards, Kai Am Donnerstag, 28. Mai 2015 08:35:08 UTC+2 schrieb Zsuzsanna P?sp?ki: > > I am trying to use the blob_dog, blob_log, blob_doh from skimage.feature > on some synthetic data. > > 1. Why do you obtain such different results for the LoG and DoG on your > sample data: http://scikit-image.org/docs/dev/auto_examples/plot_blob.html > > ? > I would expect something very similar. > > 2. I am running the following line on the attached image: blobs_log = > blob_log(image_gray, overlap = 0, min_sigma=3, max_sigma=25, num_sigma=100, > threshold=.1) > > I cannot arrive to detect precisely the size of the bigger spots (though > the range I put is enough, hence the huge spot). The increase in num_sigma > does not solve the issue. I would expect perfect results here. > > 3. Is there an easy way to visualize the N best detections? > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ralf.gommers at gmail.com Thu May 28 02:34:30 2015 From: ralf.gommers at gmail.com (Ralf Gommers) Date: Thu, 28 May 2015 08:34:30 +0200 Subject: Repository for GSoC project In-Reply-To: References: <78b496db-c70c-4a16-97a8-fa9115716280@googlegroups.com> Message-ID: On Tue, May 26, 2015 at 6:00 PM, Jaime Fern?ndez del R?o < jaime.frio at gmail.com> wrote: > > 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 at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > Thanks everyone for the feedback. After some more thinking, I have asked > Aman to start sending PRs directly against the current master of scipy. > > To differentiate new from old implementations, we'll probably just prepend > all file names with "ndi_" instead of the current "ni_". We will eventually > start removing old functionality that has been replaced, but for quite > sometime they will both coexist. This means that at some point in time, > e.g. calling `ndimage.filters.uniform_filter1d` may use the new Cython code > paths, while `ndimage.filters.minimum_filter1d` may still be using the old > C code paths. > > We will not merge anything, answering Juan's question, until we have tests > that validate that functionality. That does not mean that the functionality > will be available to end users just yet, e.g. we may merge an iterator > before any top level function that uses it. But if it does not have tests, > it does not get merged. > > If anyone has a very strong opinion against this approach, please speak up > now! > I like this. If you can keep old and new implementations side by side, it could even be a good idea to have an easy mechanism (clearly marked as temporary and only for testing) for users to switch between the two so they can test their own code with the new routines before those become the default. Ralf -------------- next part -------------- An HTML attachment was scrubbed... URL: From kwiechen1 at gmail.com Thu May 28 14:35:49 2015 From: kwiechen1 at gmail.com (Kai Wiechen) Date: Thu, 28 May 2015 11:35:49 -0700 (PDT) Subject: missing kernel or block size in equalize_adapthist In-Reply-To: References: <4b57c4b4-1f5f-4cf5-a4e6-3b14e657da2f@googlegroups.com> <903b3345-0f2d-4c5f-85bf-2c22fa4942a9@googlegroups.com> Message-ID: <814c5489-7060-45a3-af52-c2aba125cf5d@googlegroups.com> Hi Steve, yes, I would like to have finer control of block/kernel size to obtain similar results if I apply this to small test patches or the larger complete image. Is there any reason to limit ntiles_x/y to 16? Regards, Kai Am Mittwoch, 27. Mai 2015 03:17:01 UTC+2 schrieb Steven Silvester: > > Hi Kai, > > Is the issue that you want finer grain control of the block size? If so, > that would require a substantial rewrite. > > Please feel free to open an issue on Github if that is the case. > > > Regards, > > Steve > > > On May 26, 2015, at 1:58 AM, Kai Wiechen > > wrote: > > Hi Steve, > > but these parameters are limited to the range 1..16. The small test > patches are 150x150 pixels, the complete images have 1920x1448 pixels. I > have applied equalize_adapthist here with default parameters (n_tiles=8). > > Best regards, > > Kai > > On Tuesday, May 26, 2015 at 12:54:34 AM UTC+2, Steven Silvester wrote: >> >> Hi Kai, >> >> Do the `n_tiles*` arguments not meet your needs? By defining the number >> of tiles in X and Y, you are equivalently setting a block size. >> >> >> http://scikit-image.org/docs/stable/api/skimage.exposure.html#equalize-adapthist >> >> >> Regards, >> >> Steve >> >> On Monday, May 25, 2015 at 1:49:53 PM UTC-5, Kai Wiechen wrote: >>> >>> >>> >>> Hello, >>> >>> I am using equalize_adapthist after color deconvolution and some >>> morphological operations to enhance local contrast of nuclei in >>> histological images. There are different results when applying >>> equalize_adapthist when using small patches and larger size whole images >>> (left small patch, right large size whole image). The CLAHE implementation >>> in ImageJ has a 'block size' parameter and the implementation in Pixinsight >>> has a 'Kernel size' parameter to get results depending on the local >>> context. Is it possible to 'simulate' this behaviour with scikit-image? >>> >>> Best regards, >>> >>> Kai >>> >>> >>> >>> >>> >>> >>> >>> > -- > 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/2elntOClGOY/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > scikit-image... at googlegroups.com . > For more options, visit https://groups.google.com/d/optout. > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From as.zsuzsanna at gmail.com Thu May 28 14:36:12 2015 From: as.zsuzsanna at gmail.com (=?UTF-8?Q?Zsuzsanna_P=C3=BCsp=C3=B6ki?=) Date: Thu, 28 May 2015 11:36:12 -0700 (PDT) Subject: skimage.feature In-Reply-To: <3f2ef3f6-cfed-4fb9-9646-d6d7ded80388@googlegroups.com> References: <77ff627c-7334-4f28-af91-d98d9a0d12e5@googlegroups.com> <3f2ef3f6-cfed-4fb9-9646-d6d7ded80388@googlegroups.com> Message-ID: <360654a2-37c1-48a8-b18b-968e692c8034@googlegroups.com> Thanks for your message. Yes, I have tried both. I can detect all the blobs, but somehow the algorithm does not capture the size perfectly. I am wondering why there is such a difference. Actually, my intention is to use the results provided by this python code to make a comparison on real data between different algorithms (including mine). But I am not very convinced by those detections, since they do not meet with what I expect. Regards, Zsuzsanna 2015. m?jus 28., cs?t?rt?k 11:15:09 UTC+2 id?pontban Kai Wiechen a k?vetkez?t ?rta: > > Have you tried to modify the overlap or the max_sigma parameters? There is > an internal function _prune_blobs to remove overlapping signals favoring > larger ones. > > Regards, > > Kai > > > > Am Donnerstag, 28. Mai 2015 08:35:08 UTC+2 schrieb Zsuzsanna P?sp?ki: >> >> I am trying to use the blob_dog, blob_log, blob_doh from skimage.feature >> on some synthetic data. >> >> 1. Why do you obtain such different results for the LoG and DoG on your >> sample data: >> http://scikit-image.org/docs/dev/auto_examples/plot_blob.html >> >> ? >> I would expect something very similar. >> >> 2. I am running the following line on the attached image: blobs_log = >> blob_log(image_gray, overlap = 0, min_sigma=3, max_sigma=25, num_sigma=100, >> threshold=.1) >> >> I cannot arrive to detect precisely the size of the bigger spots (though >> the range I put is enough, hence the huge spot). The increase in num_sigma >> does not solve the issue. I would expect perfect results here. >> >> 3. Is there an easy way to visualize the N best detections? >> >> >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From steven.silvester at gmail.com Thu May 28 16:25:22 2015 From: steven.silvester at gmail.com (Steven Silvester) Date: Thu, 28 May 2015 15:25:22 -0500 Subject: missing kernel or block size in equalize_adapthist In-Reply-To: <814c5489-7060-45a3-af52-c2aba125cf5d@googlegroups.com> References: <4b57c4b4-1f5f-4cf5-a4e6-3b14e657da2f@googlegroups.com> <903b3345-0f2d-4c5f-85bf-2c22fa4942a9@googlegroups.com> <814c5489-7060-45a3-af52-c2aba125cf5d@googlegroups.com> Message-ID: Hi Kai, The limits were set based on the original implementation, but there is no reason they need to stay at 16. Using an arbitrary size tile is what would take a substantial rewrite. Looking at your numbers, it looks like a 13x10 block size will get you ~150x150 tile size. At the full 16x16, you'd have 120x90 pixels. Regards, Steve On Thu, May 28, 2015 at 1:35 PM, Kai Wiechen wrote: > Hi Steve, > > yes, I would like to have finer control of block/kernel size to obtain > similar results if I apply this to small test patches or the larger > complete image. Is there any reason to limit ntiles_x/y to 16? > > Regards, > > Kai > > Am Mittwoch, 27. Mai 2015 03:17:01 UTC+2 schrieb Steven Silvester: >> >> Hi Kai, >> >> Is the issue that you want finer grain control of the block size? If so, >> that would require a substantial rewrite. >> >> Please feel free to open an issue on Github if that is the case. >> >> >> Regards, >> >> Steve >> >> >> On May 26, 2015, at 1:58 AM, Kai Wiechen wrote: >> >> Hi Steve, >> >> but these parameters are limited to the range 1..16. The small test >> patches are 150x150 pixels, the complete images have 1920x1448 pixels. I >> have applied equalize_adapthist here with default parameters (n_tiles=8). >> >> Best regards, >> >> Kai >> >> On Tuesday, May 26, 2015 at 12:54:34 AM UTC+2, Steven Silvester wrote: >>> >>> Hi Kai, >>> >>> Do the `n_tiles*` arguments not meet your needs? By defining the number >>> of tiles in X and Y, you are equivalently setting a block size. >>> >>> >>> http://scikit-image.org/docs/stable/api/skimage.exposure.html#equalize-adapthist >>> >>> >>> Regards, >>> >>> Steve >>> >>> On Monday, May 25, 2015 at 1:49:53 PM UTC-5, Kai Wiechen wrote: >>>> >>>> >>>> >>>> Hello, >>>> >>>> I am using equalize_adapthist after color deconvolution and some >>>> morphological operations to enhance local contrast of nuclei in >>>> histological images. There are different results when applying >>>> equalize_adapthist when using small patches and larger size whole images >>>> (left small patch, right large size whole image). The CLAHE implementation >>>> in ImageJ has a 'block size' parameter and the implementation in Pixinsight >>>> has a 'Kernel size' parameter to get results depending on the local >>>> context. Is it possible to 'simulate' this behaviour with scikit-image? >>>> >>>> Best regards, >>>> >>>> Kai >>>> >>>> >>>> >>>> >>>> >>>> >>>> >>>> >> -- >> 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/2elntOClGOY/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> scikit-image... at googlegroups.com. >> 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/2elntOClGOY/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > scikit-image+unsubscribe at googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > -------------- next part -------------- An HTML attachment was scrubbed... URL: From vighneshbirodkar at gmail.com Fri May 29 03:18:29 2015 From: vighneshbirodkar at gmail.com (Vighnesh Birodkar) Date: Fri, 29 May 2015 00:18:29 -0700 (PDT) Subject: skimage.feature In-Reply-To: <360654a2-37c1-48a8-b18b-968e692c8034@googlegroups.com> References: <77ff627c-7334-4f28-af91-d98d9a0d12e5@googlegroups.com> <3f2ef3f6-cfed-4fb9-9646-d6d7ded80388@googlegroups.com> <360654a2-37c1-48a8-b18b-968e692c8034@googlegroups.com> Message-ID: Hello Zsuzsanna 1. According to [1], there is a constant factor difference between DoG and LoG which is ignored while computing maximas. My guess is that you will be able to get similar results using different thresholds for both methods. 2. Regarding accuracy, I am not sure you can get perfect results with these methods, since they do not go about counting individual pixels. The blob_log method uses gaussian blurring internally, and my guess is, interference from surrounding blobs is causing the difference. For you synthetic data you can use regionprops. And unless you real data has perfectly circular and single colored blobs, your measurements are going to be off anyways. 3. There is not way of doing that now. How would you define the best detection ? If accuracy is critical, we might be able to suggest other approaches, could you upload a sample of your real data set ? Or perhaps tell us what the images are like ? [1] : https://www.cs.ubc.ca/~lowe/papers/ijcv04.pdf Thanks Vighnesh On Friday, May 29, 2015 at 12:15:20 AM UTC+5:30, Zsuzsanna P?sp?ki wrote: > > Thanks for your message. > > Yes, I have tried both. I can detect all the blobs, but somehow the > algorithm does not capture the size perfectly. I am wondering why there is > such a difference. > Actually, my intention is to use the results provided by this python code > to make a comparison on real data between different algorithms (including > mine). But I am not very convinced by those detections, since they do not > meet with what I expect. > > Regards, > > Zsuzsanna > > 2015. m?jus 28., cs?t?rt?k 11:15:09 UTC+2 id?pontban Kai Wiechen a > k?vetkez?t ?rta: >> >> Have you tried to modify the overlap or the max_sigma parameters? There >> is an internal function _prune_blobs to remove overlapping signals favoring >> larger ones. >> >> Regards, >> >> Kai >> >> >> >> Am Donnerstag, 28. Mai 2015 08:35:08 UTC+2 schrieb Zsuzsanna P?sp?ki: >>> >>> I am trying to use the blob_dog, blob_log, blob_doh from skimage.feature >>> on some synthetic data. >>> >>> 1. Why do you obtain such different results for the LoG and DoG on your >>> sample data: >>> http://scikit-image.org/docs/dev/auto_examples/plot_blob.html >>> >>> ? >>> I would expect something very similar. >>> >>> 2. I am running the following line on the attached image: blobs_log = >>> blob_log(image_gray, overlap = 0, min_sigma=3, max_sigma=25, num_sigma=100, >>> threshold=.1) >>> >>> I cannot arrive to detect precisely the size of the bigger spots (though >>> the range I put is enough, hence the huge spot). The increase in num_sigma >>> does not solve the issue. I would expect perfect results here. >>> >>> 3. Is there an easy way to visualize the N best detections? >>> >>> >>> -------------- next part -------------- An HTML attachment was scrubbed... URL: From stefanv at berkeley.edu Fri May 29 12:47:37 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Fri, 29 May 2015 09:47:37 -0700 Subject: GSoC meeting notes Message-ID: <87617bwct2.fsf@berkeley.edu> 29/05/2015, VB, SW, DP Daniil has been working on an implementation of Multiblock LBP. The code is currently linked to his blog post: http://warmspringwinds.github.io/gsoc/2015/05/29/google-summer-of-code-multi-block-local-binary-patterns-implementation.-pure-python/ He will update the code with a gallery example and do a PR. So far, his MLBP implementation uses the same API as OpenCV so that it can be directly plugged in to their face recognition infrastructure for testing. We are not tied to their API, but it is convenient to use for now. DP will continue working on the rest of the object identification pipeline this week, but may return to MLBP later to convert it to Cython (if profiling shows that it is necessary). Next week's deliverables: - First step of the pre-trained OpenCV algorithm on DP?s Multiblock-LBP Start on training module in Python Deliverables: - Results of plugging MLBP into OpenCV - Pull request with MLBP + gallery example - Some updates to MLBP code requested by VB and SW Other notes: - DP will investigate whether he can get funding to attend EuroSciPy From stefanv at berkeley.edu Sun May 31 13:21:55 2015 From: stefanv at berkeley.edu (Stefan van der Walt) Date: Sun, 31 May 2015 10:21:55 -0700 Subject: Ridge filter In-Reply-To: References: Message-ID: <87d21gvf0s.fsf@berkeley.edu> Hi Juan On 2015-05-31 01:02:32, Juan Nunez-Iglesias wrote: > I found this StackOverflow Q&A through Twitter, and found the > explanation thorough and the implementation spot-on for the > problem being discussed. > > https://dsp.stackexchange.com/questions/1714/best-way-of-segmenting-veins-in-leaves This looks very interesting! I had a look at the Wikipedia page: http://en.wikipedia.org/wiki/Ridge_detection The section about variable scale detection caught my eye. Have a look at: Tony Lindeberg, "Edge detection and ridge detection with automatic scale selection" St?fan From jni.soma at gmail.com Sun May 31 04:02:32 2015 From: jni.soma at gmail.com (Juan Nunez-Iglesias) Date: Sun, 31 May 2015 18:02:32 +1000 Subject: Ridge filter Message-ID: Hi all, I found this StackOverflow Q&A through Twitter, and found the explanation thorough and the implementation spot-on for the problem being discussed. https://dsp.stackexchange.com/questions/1714/best-way-of-segmenting-veins-in-leaves However I don't recall seeing any ridge filters in skimage! Do we have any and if not, do we want to add one? Seems straightforward to implement... Juan. -------------- next part -------------- An HTML attachment was scrubbed... URL: