Mailman 3 November 2011 - scikit-image

Re: "marching squares" contribution?
by Zachary Pincus 25 Nov '11

25 Nov '11

https://github.com/scikits-image/scikits-image/pull/82 On Nov 25, 2011, at 10:23 PM, Stéfan van der Walt wrote: > Hi Zach > > On Nov 25, 2011 6:56 PM, "Zachary Pincus" <zachary.pincus(a)yale.edu> wrote: > > > > I while ago I wrote some code to extract isocontours from 2d arrays using a "marching-squares" type algorithm, with linear interpolation to get subpixel precision. (It returns a list of numpy arrays containing the coordinates of each contour...) The inner loop is in c, and it's pretty fast. Occasionally people ask me for this code, as I've mentioned making it available from time to time. > > > > Would it be a useful addition to skimage? Or a little too specific? > > This item is long overdue on our todo list and would make a very welcome addition. > > Regards > Stéfan

1 0

"marching squares" contribution?
by Zachary Pincus 25 Nov '11

25 Nov '11

Hello, I while ago I wrote some code to extract isocontours from 2d arrays using a "marching-squares" type algorithm, with linear interpolation to get subpixel precision. (It returns a list of numpy arrays containing the coordinates of each contour...) The inner loop is in c, and it's pretty fast. Occasionally people ask me for this code, as I've mentioned making it available from time to time. Would it be a useful addition to skimage? Or a little too specific? Zach

2 1

skimage @scipy India
by Emmanuelle Gouillart 14 Nov '11

14 Nov '11

Dear all, I'm going to the conference Scipy India this year, which takes place from December 4th to 7th. I'll give a tutorial on image processing similar to the one that I gave at euroscipy 2011 (http://scipy-lectures.github.com/advanced/image_processing/index.html), Compared to my earlier tutorial, I'll probably include more examples from skimage as we had a recent release with more goodies, more documentation, etc. and it's worth advertising the scikit :-D. I will also give a talk at the main conference where I plan to speak about skimage. Speaking of releases, I'd love to mention/use the new name `skimage` of the scikit during my talks. StÃ¯Â¿Â½fan, any chance that 0.4 could be released before Scipy India? There is also the possibility to have sprints at scipy india on December 6th, I'll see if some people are interested at the conference in participating to a skimage sprint. Cheers, Emmanuelle

2 1

Re: [JOB] junior developer position
by Jerome Kieffer 09 Nov '11

09 Nov '11

On Tue, 8 Nov 2011 23:51:41 -0800 Stéfan van der Walt <stefan(a)sun.ac.za> wrote: > On Nov 8, 2011 11:10 PM, "Jérôme Kieffer" <google(a)terre-adelie.org> wrote: > > I would like to advertise another position (fixed term, 3 years max) on > experiment control (using python/Qt). > > I can't read French, Weird ... sometimes the page is in french, sometimes in english with the same URL :/ Going through the portal is easier: http://www.esrf.fr/Jobs/english > so I'll have to ask: does this project also provide > the opportunity to contribute back to an open source project? By default all code ESRF produced should be available for our share-holders (mostly European countries). Making it open-source in GPL is the easiest way we found to meet this requirement (true for 95% of our projects). Some projects: https://forge.epn-campus.eu/projects/mxcube/repository http://www.tango-controls.org/ Nevertheless this contract is very peculiar (french COD, fixed term contract related to a project) is targeting at the migration from Qt3 to Qt4 of a piece of code controlling many beamlines (7 at ESRF, but also in some other places) So contributing back to open-source projects is possible (within this COD) only for closely related open-source project on which MXCube depends like PyQt, PyQwt, Qwt, Tango, PyMca, ... Sorry, that's probably not the answer you expected :-/ -- Jerome Kieffer PS: Software Engineer (m/f), short-term contract of 18 months minimum and 36 months maximum time-limited contract THE FUNCTION As part of the general ESRF beamline portfolio renewal, the Structural Biology Group will create a new suite of state of the art beamlines for Macromolecular Crystallography (MASSIF). The facility will be built on ID30 and will comprise three beamlines, one with micro-focus capability. All beamlines will be equipped with automated sample changers and advanced sample evaluation instrumentation and software. You will participate in the enhancement of our Python/Qt based beamline control application, mxCuBE required for the implementation of advanced sample evaluation, in two distinct phases : • Adaptation of the current product to the latest developments of our sample evaluation and diffraction data processing platform (EDNA) as well as the development of the communication bothways to our laboratory information system (ISPYB) via webservices and implementation of more complicated sample characterisation and data collection strategies • Migration of mxCuBE to the new ESRF platform for beamline control applications The mission also involves deployment, support and maintenance tasks to allow a continuous usage of the software by our users during their experiments. Together with the opportunity to extend your knowledge of data acquisition you will work as a member of a team of scientists and developers dedicated to bringing the most advanced sample evaluation and data collection techniques to the European Macromolecular Crystallography Community. mxCuBE is also used by other synchrotrons within Europe and you’ll have therefore to enlarge your interactions to developers outside ESRF. This mission will be considered as finished when the migration of MxCuBE to the new platform is completed and all functionality is tested and stable. QUALIFICATIONS AND EXPERIENCE You should have a higher university degree (MSc, DESS, Master, Diploma, Licenciado, Laurea or equivalent) in Software Engineering. Professional skills with Python programming and the Qt graphical toolkit (PyQt) are mandatory. Knowledge of Javacript and web application development would be an asset. Working experience with git or a similar revision control system would be appreciated. Ideally, you should have a genuine interest in distributed systems, scientific instrumentation software and associated techniques. You should be highly motivated and capable of integrating into the ESRF's multidisciplinary and international staff; ability to work within a dynamic international collaboration is essential. Ref. COD/MXCUBE - Deadline for returning application forms: 18/11/2011

1 0

Re: [JOB] junior developer position
by Jérôme Kieffer 09 Nov '11

09 Nov '11

Hi all, I would like to advertise another position (fixed term, 3 years max) on experiment control (using python/Qt). http://esrf.profilsearch.com/recrute/fo_annonce_voir.php?id=72 -- Jérôme Kieffer

2 1

[JOB] junior developer position
by Emmanuelle Gouillart 08 Nov '11

08 Nov '11

Dear all, a 1-year Python (and ImageJ) developer position on 3-D image processing is available in my team, hopefully involving some contributions to skimage! See the description below, and also http://www.svi.cnrs-bellevue.fr/wikimedia/index.php/EDDAM_developer_position Please spread the word! Cheers, Emmanuelle ***************** Hiring a junior developer on 3-D image processing in the joint lab CNRS/Saint-Gobain (Paris) Description of the position --------------------------- The goal of this 1-year project is to build a user-friendly and good-quality 3-D image processing package from the miscellaneous codebases written by different research teams working together on X-ray tomography. In a first phase, you will strongly interact with the researchers of the different teams in order to list and collect the different algorithms used by the teams (including a few state-of-the-art algorithms used by specialists on image processing collaborating with the teams). You will integrate these algorithms (starting with those of most common interest) into a common package, taking special care of usability (documentation, installation, ...) and robustness of the code (testing, handling of different file formats). The package will consist of a Python package and a set of ImageJ routines. For easier maintenance, some algorithms of general interest will be contributed instead to the open-source Python scikits-image ``skimage``. Scientific context ------------------ The French EDDAM project (http://www.svi.cnrs-bellevue.fr/wikimedia/index.php/EDDAM) focusses on the ultrafast 3-D X-ray imaging of amorphous materials under mechanical load or thermal treatment. Thanks to state-of-the-art developements of synchrotron X-ray tomography, it is now possible to image the evolution of materials at timescales below 1s, giving unprecendented insights into the transformation of materials. Such experiments typically produce huge datasets, for which efficient automated image processing methods are required. Requirements ------------ * Good abilities for team working. Objective-driven mindset. * Knowledge of a programming language and some scientific computing. Knowledge of one of the languages/sotfwares used by the team (Python, ImageJ + Java, Matlab) is a plus. * Interest for quality assurance and best practices in software development: documentation, testing, version control. * Interest and curiosity on image processing. A previous experience on image processing or a mathematical-oriented mindset is a plus. Practical aspects ----------------- The position is available from February 2012. The salary depends on the level of experience of the candidate, from 1650 euros (net salary) per month (master's degree) to 1930 euros per month (PhD degree). If you are finishing a master's degree and you are looking for an internship, it is also possible to start with an internship before taking the 1-year position. Your employer will be the CNRS (French National Research Center), and you will be located at the joint unit CNRS/Saint-Gobain (http://www.svi.cnrs-bellevue.fr), in Aubervilliers, close to Paris (France). Candidates should send a CV with cover letter to Emmanuelle Gouillart at emmanuelle.gouillart(a)nsup.org.

2 1

documentation marathon :-)
by Emmanuelle Gouillart 02 Nov '11

02 Nov '11

Hello, as I was browsing through the scikit, I started to make a list of functions for which we don't have examples (neither in their docstring, nor an example script for the documentation/example gallery). I copy it here, since it seems a priority to me that we have more examples in the documentation. I was thinking of writing an example of edge detection wih canny and sobel, and completing their docstrings while I'm at this. Any other volunteer to make this list shrink :-)? Cheers, Emmanuelle ------------ feature: hog filter: median_filter, canny, sobel, prewitt, wiener graph: shortest_path, MCP morphology: * structuring elements in selem.py don't have examples in docstrings. * In grey.py, Tony added recently examples in all the docstrings -- it'd be great to have also a few examples for the gallery * label doesn't have examples transform: probabilistic_hough, homography, integral_image, integrate radon and iradon do not have examples in docstrings (but they do have a nice example in the gallery)

3 3

Re: Renaming / 0.4 timeline
by Stéfan van der Walt 01 Nov '11

01 Nov '11

Hi all, 2011/10/18 Stéfan van der Walt <stefan(a)sun.ac.za>: > Tue Oct 18th to Sat Oct 22nd : File and merge PRs of work-in-progress > made against the current repo > Sun Oct 23rd : Rename the repository, update documentation, test > Mon Oct 24th to Fri Oct 28th : Apply bugfixes, add features, etc. > required for 0.4, including polishing the Debian package > Oct 29th to 30th : Release v0.4 Is everybody ready for the 0.4 release? Any last minute patches you want to get in? Regards Stéfan

1 0