Mailman 3 April 2014 - scikit-image

Re: skimage and opencv
by Emmanuelle Gouillart 22 Sep '16

22 Sep '16

Hi FranÃ¯Â¿Â½ois, that's an excellent question, and not a troll :-). Opencv is a very powerful library, but it focuses primarily on computer vision (feature detection and extraction, classification, ...), as opposed to image processing in general (with other tasks such as denoising, segmentation, ...). The other big difference is that skimage builds on numpy ndarrays, and uses the full power of the numpy API (including of course the basic facilities for processing arrays as images that come with numpy), as well as some of scipy functions (you could have added scipy.ndimage to your list -- a few functions in skimage are wrappers around scipy.ndimage, that exist for the sake of completeneness). One important consequence is that algorithms working for 3-d or even n-d images can be easily implemented in 3-d/n-d in skimage, whereas opencv is restricted to 2-D images (as far as I know). Thanks to the use of numpy arrays, the API of skimage is also quite pleasant for a numpy user, more than the API of opencv. A related difference is that skimage is written in python and cython, whereas opencv is a C++ library. The two libraries attract a different crowd of developers, and a Python/Cython toolkit based on numpy arrays is easier to develop and maintain inside the Scientific Python ecosystem. I'm sure that other devs/users will have things to add to this discussion! Cheers, Emmanuelle On Thu, Dec 27, 2012 at 02:06:08PM -0800, FranÃ¯Â¿Â½ois wrote: > Hi users and devs, > It came to my knowledge that another python library (based on C++ and C > codes) for image processing exists too : opencv > I understand that numpy intregrates some basic features and we need some > advanced features but I have the feeling that skimages is redoundant with > opencv in some ways. > What's the position of skimage about that? (Don't read this question as a > troll but like a real question). > I mean that similar features exist in both. Would not be possible to > reuse/integrate opencv or merge? what's the reason for keeping them apart? > My observation is there is 4 libraries to manipulate images: > * PIL > * numpy > * skimages > * opencv > That's a lot. > Cheers,

3 2

Re: Contribution: circles and hough transform
by François 18 Jun '14

18 Jun '14

Hi, FYI, I plan to work on the detection of ellipses quite soon. Stefan indicated to me a reference which seems fairly simple to implement. Cheers, François.

3 3

Re: "Cythonize" not working on OSX?
by Stéfan van der Walt 06 Jun '14

06 Jun '14

On Mon, 03 Feb 2014 16:44:07 +1100, Juan Nunez-Iglesias wrote: > I didn't know this was a possible course of action! =) I have no idea how > many people are affected, or how quick a fix can be... Is it worth > exploring a bit more before reversing it? I actually have very little idea > about the entire build process so I can't offer much insight here... It broke the build on my system too (perhaps it does not check correctly for old versions), so my preference is to revert and have another look at it. StÃ©fan

3 2

Re: Potential contribution: subpixel cross correlation
by Juan Nunez-Iglesias 29 Apr '14

29 Apr '14

On Sun, Apr 27, 2014 at 10:53 AM, Stéfan van der Walt <stefan(a)sun.ac.za>wrote: > > Perhaps Juan can answer this one--he is our most vocal 3D proponent! > Haha, 3D is not the same as 2D+c! I'm actually also used to grayscale EM images, but in 3D. For that case, I think the generalisation would be straightforward. For 2D+c, I have no idea! In the meanwhile, getting a version up and running with tests in 2D is > a great start. > Or 3D grayscale. Just a suggestion. ;) Juan.

1 0

Re: Real-time interaction forum
by Stuart Mumford 27 Apr '14

27 Apr '14

Hello, > https://github.com/Cadair/williemodules it is very basic but it works. > > That's really neat! > There are at least two quite major feature missing at the moment, the main one is rate limiting the notifications! The second is optionally turning off the notification when the user is online. Also from my limited testing I think that all the emails are being duplicated! > Thanks, that's very kind of you. I think a lot of people would > appreciate having an IRC channel. Does the logger update a page in > real-time so that a person could "watch" IRC via the web? > Yes, it has a js front end that updates in "real time" there is a variable delay, but it will update. Currently what the logger is missing is a decent way of scrolling through the history more than about 100-200 messages in the past (the js seems to go loopy). All the messages are in a MySQL database so I want to code up a better flask-based front end to it, but you know time is always a problem. (I also have a lovely little idea for integrating the logger and the notification module.) Stuart

1 0

Re: Real-time interaction forum
by Stéfan van der Walt 27 Apr '14

27 Apr '14

Hey, Stuart On Sat, Apr 26, 2014 at 9:10 PM, Stuart Mumford <stuart(a)mumford.me.uk> wrote: > Partly motivated by this conversation, and partly by my need to experiment, > I have written a module for the IRC bot we use in SunPy > (https://github.com/embolalia/willie/) so that you can tell it to send you > pushover (http://pushover.net) or email notifications when your nick is > mentioned in a channel the bot is in. My code so far is here: > https://github.com/Cadair/williemodules it is very basic but it works. That's really neat! > I would be happy to run a bot and a logger as already mention if you guys > decide to set up a IRC channel, so this would at least give you > notifications. Thanks, that's very kind of you. I think a lot of people would appreciate having an IRC channel. Does the logger update a page in real-time so that a person could "watch" IRC via the web? Stéfan

1 0

Re: Potential contribution: subpixel cross correlation
by Stéfan van der Walt 27 Apr '14

27 Apr '14

On Sun, Apr 27, 2014 at 2:48 AM, Michael Sarahan <msarahan(a)gmail.com> wrote: > Your example highlights a critical issue: handling color data. I've only > ever played with grayscale data (electron microscope images). I tested my > code with your example image, and it falls over. Do you have any tips for > what special handling I might need to do (convolutions, finding maxima)? I > can just grayscale the input images, but that doesn't feel like an optimal > solution. Perhaps Juan can answer this one--he is our most vocal 3D proponent! In the meanwhile, getting a version up and running with tests in 2D is a great start. Stéfan

1 0

Re: Potential contribution: subpixel cross correlation
by Michael Sarahan 27 Apr '14

27 Apr '14

Hi Stefan, Your example highlights a critical issue: handling color data. I've only ever played with grayscale data (electron microscope images). I tested my code with your example image, and it falls over. Do you have any tips for what special handling I might need to do (convolutions, finding maxima)? I can just grayscale the input images, but that doesn't feel like an optimal solution. Thanks, Mike On Sat, Apr 26, 2014 at 4:09 PM, Stéfan van der Walt <stefan(a)sun.ac.za>wrote: > On Sat, Apr 26, 2014 at 10:50 PM, Michael Sarahan <msarahan(a)gmail.com> > wrote: > > Based on the docs, it > > looks like that's also true for scikit-image? > > Here's what I meant: > > > http://nbviewer.ipython.org/urls/dl.dropbox.com/s/dmqwusnh9hlk505/find_shif… > > This is very slow, though, and a bit of a hack, so I'm happy to > include this algorithm (if, as Tony mentioned, the license checks > out). > > Thanks! > Stéfan > > -- > You received this message because you are subscribed to the Google Groups > "scikit-image" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to scikit-image+unsubscribe(a)googlegroups.com. > For more options, visit https://groups.google.com/d/optout. >

1 0

Re: Potential contribution: subpixel cross correlation
by Stéfan van der Walt 26 Apr '14

26 Apr '14

On Sat, Apr 26, 2014 at 10:50 PM, Michael Sarahan <msarahan(a)gmail.com> wrote: > Based on the docs, it > looks like that's also true for scikit-image? Here's what I meant: http://nbviewer.ipython.org/urls/dl.dropbox.com/s/dmqwusnh9hlk505/find_shif… This is very slow, though, and a bit of a hack, so I'm happy to include this algorithm (if, as Tony mentioned, the license checks out). Thanks! Stéfan

1 0

Re: Potential contribution: subpixel cross correlation
by Michael Sarahan 26 Apr '14

26 Apr '14

I've emailed the original author (Manuel Guizar) and asked about licensing. I'll make sure it's OK to bring under the scikit-image license before I do a PR. Best, Mike On Sat, Apr 26, 2014 at 1:58 PM, Tony Yu <tsyu80(a)gmail.com> wrote: > >> On Sat, Apr 26, 2014 at 12:02 PM, Stéfan van der Walt <stefan(a)sun.ac.za>wrote: >> >>> Hi Mike >>> >>> On Sat, Apr 26, 2014 at 8:24 PM, Mike Sarahan <msarahan(a)gmail.com> >>> wrote: >>> > >>> http://www.mathworks.com/matlabcentral/fileexchange/18401-efficient-subpixe… >>> >>> > This sounds interesting, but I'm a bit worried by the license. Most posts > on the File Exchange have a BSD license, but this particular contribution > has "No BSD License" written next to it. I'm not sure what the protocol is > for derivative works, but it's something to keep in mind. > > Best, > -Tony > > -- > You received this message because you are subscribed to the Google Groups > "scikit-image" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to scikit-image+unsubscribe(a)googlegroups.com. > For more options, visit https://groups.google.com/d/optout. >

1 0