Mailman 3 November 2013 - scikit-image

Re: scikit-image paper
by Johannes Schönberger 20 Nov '13

20 Nov '13

So, how should we get started with this? Any roadmap / plan? Am 19.11.2013 um 05:49 schrieb Tony Yu <tsyu80(a)gmail.com>: > > > > On Mon, Nov 18, 2013 at 8:50 PM, Marianne Corvellec <marianne.corvellec(a)ens-lyon.org> wrote: > Hello all, > > What an exciting project! > > > On Monday, November 18, 2013 3:59:22 AM UTC-5, Stefan van der Walt wrote: > Let's get started on some content! > > I think we would do well to emphasize the following in the paper: > > - Pythonic API > - Building block for reproducible research, use in academic world (we > need a list of publications of work that used skimage) > - Educational aspects (skimage.novice) > - Google Summer of Code contributions > - Use in industry (do we have examples here?) > - ... other ideas? > > I believe it would be fitting to cite Dharhas Pothina's recent work. > Stéfan (and possibly others) know more about it than I do, but it could be considered both academic and industrial. > > I believe Tony S Yu would have lots to share as well. :) > > Thanks for pinging me. > > Unfortunately, the current project I'm working on isn't public (at least when I last checked). There's another enthought project that is public: I'll ask around to see if it's OK to make a direct reference to the project. > > -T > > -- > You received this message because you are subscribed to the Google Groups "scikit-image" group. > To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image+unsubscribe(a)googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out.

2 1

Re: Advice on improving a analysis on images of nanoparticles
by Guillaume Gay 19 Nov '13

19 Nov '13

Hi all, I'm currently working with cell nuclei segmentation (in 3D) with fluorescence imagery. I'm know using quite a pannel of sckimage routines to segment those out (local contrast enhancement, threshold otsu, etc.)?. Would you be interested if I turned that into an example (maybe something that could be chunked in several pieces?). Best, Guillaume On 20/11/2013 07:13, Stéfan van der Walt wrote: > > On 20 Nov 2013 07:33, "Juan Nunez-Iglesias" <jni.soma(a)gmail.com > <mailto:jni.soma@gmail.com>> wrote: > > > > I'm guessing you are applying label() directly to your image, which > is not the right way to use it. > > Looks like isolating objects is a common problem. Time for another > gallery example? We already have the coins, but perhaps something more > realistic with some biological data. > > Stéfan > > -- > You received this message because you are subscribed to the Google > Groups "scikit-image" group. > To unsubscribe from this group and stop receiving emails from it, send > an email to scikit-image+unsubscribe(a)googlegroups.com. > For more options, visit https://groups.google.com/groups/opt_out.

1 0

Re: Advice on improving a analysis on images of nanoparticles
by Stéfan van der Walt 19 Nov '13

19 Nov '13

On 20 Nov 2013 07:33, "Juan Nunez-Iglesias" <jni.soma(a)gmail.com> wrote: > > I'm guessing you are applying label() directly to your image, which is not the right way to use it. Looks like isolating objects is a common problem. Time for another gallery example? We already have the coins, but perhaps something more realistic with some biological data. Stéfan

1 0

Re: Advice on improving a analysis on images of nanoparticles
by Evelyn Liu 19 Nov '13

19 Nov '13

Hi Juan, On Tuesday, November 19, 2013 9:39:47 PM UTC-5, Juan Nunez-Iglesias wrote: > > Is the goal only to count particles? In that case, I think a local > thresholding (threshold_adaptive) would work on all these images. Then, > just do a labelling (scipy.ndimage.label) and draw a histogram of particle > sizes. You'll get a sharp peak around the true particle size, with bigger > peaks for clumps > I'd like to use scikit to plot the size distribution histogram of particles in an image, which is similar with Adam's. I tried scipy.ndimage.measurements.label(image), which I thought would give an array about particle sizes. However, the output array is with all 1, obviously nothing about size. I must get something wrong...So which function should i call for the size distribution? Thanks Juan!

2 1

Re: Surface Curvature from Range Images (or Depth Maps)
by Brickle Macho 19 Nov '13

19 Nov '13

Hi StÃ¯Â¿Â½fan and others, As requested. Depth maps provide data on non-flat surfaces. Curvature is the amount by which a object/surface deviates from being flat. Using differential geometry, we can find the compute the curvature of the surfaces in an image. Due to the structure of a depth map (essentially a Monge Patch see: http://mathworld.wolfram.com/MongePatch.html), numerical computation of the surface curvature map only requires the first and second partial derivative estimates at each surface point and can be estimated using np.gradient() function. Here is example output form the provided code: http://i.imgur.com/kyMaR1F.jpg If interested, slightly more detail provided after my signature block. I have no ego over this, so if not appropriate to skimage that is ok. Heck, someone else can take control and include this into skimage, but I suspect like me, everyone is busy with life and other side projects. Michael. -- What follows is a more detailed version of above. Apologies if some of the following is obvious some. In my current research, my focus is on depth maps or range images. Others in our research ground are using point clouds. The benefit of depth maps is that they are basically a structured point cloud but allow me to use common 2D image processing algorithms on the depth maps, and not worry about a specific point cloud library or algorithms. For example applying some simple morphological operations (from skimge) I have been segment objects at varying depths, i.e segment objects that standout. A depth map is gray-scale image which encodes the distance for a certain view or perspective. Depth maps record the first surface seen and (obviously) cannot display information about occluded surfaces, surfaces seen or refracted through transparent object or reflected in mirrors. I also have an aligned with the color view of the same scene, but not directly relevant to this post. Curvature is the amount by which a object/surface deviates from being flat. Curvature can be used to describe a surface. Either using the curvature pixel value itself as the feature or calculating one of eight surface primitives are: peak, pit, ridge, valley, flat, minimal, saddle ridge, and saddle valley surfaces. In the code provided in the original post I just calculate the the curvature value of each pixel, resulting in a curvature feature map. Technically the code provided can be applied to any grey-scale image, but really only makes sense when applied to range iamges. Depth maps provide data on non-flat surfaces. Differential geometry can be used to process and analyse data on non-flat surfaces. Using differential geometry, we can find the compute the curvature of the surfaces in an image. A depth or range value at a point (x, y ) is given by a single-valued function z(x,y), essentially a Monge Patch see: http://mathworld.wolfram.com/MongePatch.html. Because of this the Gaussian and mean curvatures are only related to the first and second partial derivatives of the function z with respect to x and y. Thus the numerical computation of the surface curvature map only requires the first and second partial derivative estimates at each surface point. These can be estimated using np.gradient() function. Once we have these estimates, it is simply a mater of substituting into the equations from the various papers. Here is the code applied to a range image from the Middlebury data set (http://vision.middlebury.edu/stereo/data/). The linked image shows he original range image, mean and guassian curvature, and simplified mean and guassian curvature. The RGB imagery is included for completeness but is not used in the calculation. Curvature Example: http://i.imgur.com/kyMaR1F.jpg On 19/11/2013 11:37 pm, StÃ¯Â¿Â½fan van der Walt wrote: > Dear Michael > > On Wed, Nov 13, 2013 at 3:19 PM, Brickle Macho <bricklemacho(a)gmail.com> wrote: >> I have implemented mean and gaussian curvature methods in python and thought >> they may be appropriate for inclusion to skimage. An outline of the code is >> provided below. It assumes a surface defined as a function of two >> coordinates, e.g. z = Z(x, y). The curvature calculations are from the >> following two papers > Thank you for your post, and sorry for not responding earlier: I was > hoping someone more knowledgeable would jump in. Since I am not that > person, would you mind explaining to me the gist of this work, and > what it aims to do, perhaps with some example images? > > Thanks! > StÃ¯Â¿Â½fan >

1 0

Re: Advice on improving a analysis on images of nanoparticles
by Adam Hughes 19 Nov '13

19 Nov '13

Thanks for the help Stefan. Here is a direct link that anyone can view: http://www.scribd.com/doc/185189653/Scikit-Test-Notable To download without an account, I am not familiar with any hosting solution, but if you guys have any recommendations I'd love to hear them. Thank you for you help, I will test out the methods you suggested On Sunday, November 17, 2013 4:27:12 AM UTC-5, Stefan van der Walt wrote: > > Hi Adam > > On Fri, Nov 15, 2013 at 7:02 PM, Adam Hughes <hughes...(a)gmail.com<javascript:>> > wrote: > > The images are in a .pdf can be downloaded directly here. (Just a hair > too > > big to attach) > > This download requires a login--could you please provide us with a > direct link / HTML page showing the images? > > > LOWCONTRAST - Can the contrast in these images be enhanced automatically > in > > SKI? > > We have several histogram equalization algorithms, including Contrast > Limited Histogram Equalization, which should do the trick. > > > NONCIRC - Particles appear non-circular due to stigmation offset in > > microscope. Is it possible to reshape them/make them more circular? > > It depends on what you mean by "make them more circular". We have > geometric transformation utilities in `skimage.transform`. > > > WARPED - Images that have artifacts, or uneven contrast, due to > aberrations > > in SEM beam during imaging. I'm especially interested in removing > uneven > > contrast. > > Localized contrast equalization in the `skimage.filter.rank` module might > help. > > > WATERSHED - These images have overlapping AuNPs, and I had hoped that > SKI's > > watershedding routines might help disentangle them. The watershed > > segmentation guide indicates that there are several ways to approach > this > > problem. > > See Juan's post here: > https://groups.google.com/d/msg/scikit-image/4z-hPiFFDj8/2gIDMxfGrU4J > > Regards > Stéfan >

2 1

Surface Curvature from Range Images (or Depth Maps)
by Brickle Macho 19 Nov '13

19 Nov '13

I have implemented mean and gaussian curvature methods in python and thought they may be appropriate for inclusion to skimage. An outline of the code is provided below. It assumes a surface defined as a function of two coordinates, e.g. z = Z(x, y). The curvature calculations are from the following two papers * Kurita, T., & Boulanger, P. (1992). Computation of Surface Curvature from Range Images Using Geometrically Intrinsic Weights. In /MVA/ (pp. 389-392). * Zhao, C., Zhao, D., & Chen, Y. (1996, August). Simplified Gaussian and mean curvatures to range image segmentation. In /Pattern Recognition, 1996., Proceedings of the 13th International Conference on/ (Vol. 2, pp. 427-431). IEEE. If appropriate let me know and I will read the contribution/development documentation, make the code more robust and submit formally. Regards, Michael. -- Here is a basic outline of the code (H and K are used as they appear common terms used in the literature) : ---------------BEGIN CODE-------------- import numpy as np Zy, Zx = np.gradient(Z) Zxy, Zxx = np.gradient(Zx) Zyy, _ = np.gradient(Zy) # Mean Curvature - equation (3) from Kurita and Boulanger (1992) paper # See also Surface in 3D space, http://en.wikipedia.org/wiki/Mean_curvature H = (1 + (Zx ** 2)) * Zyy + (1 + (Zy ** 2)) * Zxx - 2 * Zx * Zy * Zxy H = H / ((2 * (1 + (Zx ** 2) + (Zy ** 2))) ** 1.5) # Gaussian Curvature - equation (4) from Kurita and Boulanger (1992) paper K = (Zxx * Zyy - (Zxy ** 2)) / ((1 + (Zx ** 2) + (Zy **2)) ** 2) # Simplified Mean Curvature - equation (3) from Zhao et.al (1996) paper H = Zxx + Zyy # Simplified Gaussian Curvature - equation (3) from Zhao et.al (1996) paper K = Zxx * Zyy - (Zxy ** 2) ---------------END CODE--------------

2 1

Re: scikit-image paper
by Marianne Corvellec 18 Nov '13

18 Nov '13

Hello all, What an exciting project! On Monday, November 18, 2013 3:59:22 AM UTC-5, Stefan van der Walt wrote: > > Let's get started on some content! > > I think we would do well to emphasize the following in the paper: > > - Pythonic API > - Building block for reproducible research, use in academic world (we > need a list of publications of work that used skimage) > - Educational aspects (skimage.novice) > - Google Summer of Code contributions > - Use in industry (do we have examples here?) > - ... other ideas? > I believe it would be fitting to cite Dharhas Pothina's recent work. Stéfan (and possibly others) know more about it than I do, but it could be considered both academic and industrial. I believe Tony S Yu would have lots to share as well. :) Marianne

2 1

Re: scikit-image paper
by Adam Hughes 18 Nov '13

18 Nov '13

On Monday, November 18, 2013 3:59:22 AM UTC-5, Stefan van der Walt wrote: > On Mon, Nov 18, 2013 at 10:47 AM, Juan Nunez-Iglesias > <jni....(a)gmail.com <javascript:>> wrote: > > If we are going to worry about JIF, PLOS ONE has 3.7, and does accept > papers > > from all disciplines. (You will also find a Scholarly Kitchen post about > how > > a JIF of 3.7 is a sign that PLOS and open access are doomed.) > > Thanks, I won't take Scholarly Kitchen too seriously then. I know > that impact factors are a lousy way (by itself) to determine where to > publish, but that's a whole other kettle of fish. > > > Currently my two top candidates then are JMLR MLOSS and PLOS ONE. (The > > latter does charge $1300 US per article.) > > Open access fees should not pose too much of a problem: > > http://library.sun.ac.za/English/services/oa/Pages/su-oafund.aspx > > Let's get started on some content! > > I think we would do well to emphasize the following in the paper: > > - Pythonic API > - Building block for reproducible research, use in academic world (we > need a list of publications of work that used skimage) > - Educational aspects (skimage.novice) > - Google Summer of Code contributions > - Use in industry (do we have examples here?) > It may be non-disclosable, but I know that Enthought may have some industrial examples. > - ... other ideas? > > Stéfan >

1 0

Re: SLIC Module in SciKit
by Jeremy Dunne 18 Nov '13

18 Nov '13

Yep I'd say so. Cheers Stefan Regards. Jeremy. On Monday, 18 November 2013 16:50:49 UTC, Stefan van der Walt wrote: > > On Mon, Nov 18, 2013 at 6:49 PM, Jeremy Dunne <jdun...(a)gmail.com<javascript:>> > wrote: > > Well that seems to have solved it...thank you very much....atleast I > have a > > new error ..... > > Great! The new error seems unrelated to skimage. > > Stéfan >

1 0