Mailman 3 October 2014 - scikit-image

MultiImage, converse_memory and tifffile plugin
by jim＠rybarski.com 12 Oct '14

12 Oct '14

So recently I had to stop using MultiImage to open up a large TIFF stack as I kept getting memory errors, even when using converse_memory=True (it ends up loading the entire file into memory regardless). I installed the tifffile module and used that instead, as it has the ability to open the file but load only one image at a time. I think it would be nice if MultiImage was able to do this automatically (or by getting a plugin argument), especially since the behavior of the TiffFile class is conceptually very similar to MultiImage. Thoughts?

2 1

Handy Bash Function for Creating Branches
by Steven Silvester 11 Oct '14

11 Oct '14

Hi all, I added the following to my .bashrc for creating new branches: function gn() { git checkout -b "$1" upstream/master } I use it as follows: $ gn temp2 Branch temp2 set up to track remote branch master from upstream. Switched to a new branch 'temp2' This saves me from forgetting to branch against upstream master all the time. Regards, Steve

3 2

RGB Value Confusion
by Michael Alonge 09 Oct '14

09 Oct '14

Hello, Maybe this is quite trivial, but I am a little confused about the RGB values represented in my numpy arrays. I get floats between zero and one, such as (0.45803295125156029, 0.10643323197708233, 0.18076931323028619) (this should be a slightly red color). Is there a way for me to convert this to normal RGB values? My understanding is that they should be whole integers. Thank you, Michael Alonge

2 1

ellipse-shaped structuring element for morphology
by Pratap Vardhan 07 Oct '14

07 Oct '14

Here's a ellipse-shaped structuring element for morphology operations In [131]: def ellipse(w, h, dtype=np.uint8): """ Generates a flat, ellipse-shaped structuring element. Parameters ---------- w : int The width of ellipse h : int The height of ellipse Other Parameters ---------------- dtype : data-type The data type of the structuring element. Returns ------- selem : ndarray A structuring element consisting only of ones, i.e. every pixel belongs to the neighborhood. """ X, Y = np.meshgrid(range(-w, w + 1), range(-h, h + 1)) return np.array(((h*X) ** 2 + (w*Y) ** 2) <= (w*h) ** 2, dtype=np.uint8) In [129]: ellipse(3, 3) Out[129]: array([[0, 0, 0, 1, 0, 0, 0], [0, 1, 1, 1, 1, 1, 0], [0, 1, 1, 1, 1, 1, 0], [1, 1, 1, 1, 1, 1, 1], [0, 1, 1, 1, 1, 1, 0], [0, 1, 1, 1, 1, 1, 0], [0, 0, 0, 1, 0, 0, 0]], dtype=uint8) In [130]: ellipse(5, 3) Out[130]: array([[0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0], [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0], [0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1], [0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0], [0, 0, 1, 1, 1, 1, 1, 1, 1, 0, 0], [0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0]], dtype=uint8) Can this be taken as it is or certain approximation formulas need to be used? Any alternative way skimage uses as of now? If not, can this go as a PR?

2 2

Re: Memory Error When Converting to LAB Color Space.
by Johannes Schönberger 06 Oct '14

06 Oct '14

> With that version of Python, you cannot access even a fraction of that memory. He should be able to access a “fraction" of the memory, though ;-) Johannes Schönberger On Oct 6, 2014, at 3:50 PM, Stéfan van der Walt <stefan(a)sun.ac.za> wrote: > On Oct 6, 2014 8:28 PM, "Michael Alonge" <malonge11(a)gmail.com> wrote: > > > > Thanks so much for the help. > > > > I am processing with 32 GB of RAM and a 64-bit operating system. That being said, I am running the 32 bit version of python. > > With that version of Python, you cannot access even a fraction of that memory. Can you try the 64-bit version? > > 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

Memory Error When Converting to LAB Color Space.
by Michael Alonge 06 Oct '14

06 Oct '14

Hello, This is my first post in this forum. Thank you in advance for any and all help. I have an image that when initially loaded in RGB yields a memory error when converted to LAB via color.rgb2lab(image). I get the following error: >>> lab_image = color.rgb2lab(image) Traceback (most recent call last): File "<pyshell#1>", line 1, in <module> lab_image = color.rgb2lab(image) File "C:\Python33\lib\site-packages\skimage\color\colorconv.py", line 801, in rgb2lab return xyz2lab(rgb2xyz(rgb)) File "C:\Python33\lib\site-packages\skimage\color\colorconv.py", line 523, in rgb2xyz arr = _prepare_colorarray(rgb).copy() MemoryError The picture itself has the following specs: File Type: .PNG File size: 20 MB Resolution: 72 ppi image.shape = (5184, 3456, 3) I am working on a windows 7 machine with python 3.3.4. I am an intermediate python programmer, and have no idea where to start when it comes to memory allocation. Any and all help is appreciated, Thank you, Michael Alonge

2 4

Re: Memory Error When Converting to LAB Color Space.
by Stéfan van der Walt 06 Oct '14

06 Oct '14

On Oct 6, 2014 8:26 AM, "Tony Yu" <tsyu80(a)gmail.com> wrote: > > That was more of a test of re-allocation of memory (i.e. let's make sure there's no memory leaks), but there may be issues with total available memory. The arrays are stored in "out", so my aim was to test total memory allocation (which seems fine). I can't imagine that we are making so many copies, but I could be wrong. Best is to run a memory profile. Stéfan

1 0

Re: robust ring (and circle) detection algorithm
by Eldad Afik 06 Oct '14

06 Oct '14

Hello, Unfortunately I never got to revising the code for a pull request, nor documenting it as I would have liked. A Cython code is available at ridge-directed-ring-detector <https://github.com/eldad-a/ridge-directed-ring-detector> where you can also find a usage example in the format of an ipython-notebook; a static view is available here <http://nbviewer.ipython.org/github/eldad-a/ridge-directed-ring-detector/blo…> . For those who are interested, the manuscript has been updated (version v2) and is available at arXiv:1310.1371 <http://arxiv.org/abs/1310.1371> . In case there is interest from the community, I can try to modify it to conform to the scikit-image standards, based on your comments. Congratulations for the new scikit-image paper! Best wishes, Eldad On Thursday, 13 March 2014 23:34:32 UTC+2, Stefan van der Walt wrote: > > Hi Eldad > > On Sun, Mar 9, 2014 at 3:28 PM, Eldad Afik <eldad...(a)gmail.com > <javascript:>> wrote: > > I summarised the work in a manuscript titled: > > "Robust and highly performant ring detection algorithm for 3d particle > > tracking using 2d microscope imaging" (a pre-print is available at > > arXiv:1310.1371). > > NB, it was written minded at potential users and applications rather > than > > computer scientists, which I am not. > > Shall we start working on a pull request with a proof of concept > illustration? > > Stéfan >

1 0

Re: Memory Error When Converting to LAB Color Space.
by Michael Alonge 06 Oct '14

06 Oct '14

It looks like it worked. I got the following output: Allocating # 0 Allocating # 1 Allocating # 2 Allocating # 3 Allocating # 4 Thank you, Mike Alonge On Thursday, October 2, 2014 3:50:17 PM UTC-7, Stefan van der Walt wrote: > > Hi Michael > > On Thu, Oct 2, 2014 at 7:17 PM, Michael Alonge <malo...(a)gmail.com > <javascript:>> wrote: > > I have an image that when initially loaded in RGB yields a memory error > when > > converted to LAB via color.rgb2lab(image). I get the following error: > > Just as a basic memory size check, can you please let me know whether > the following script executes successfully? > > import numpy as np > shape = (5184, 3456, 3) > > out = [] > for i in range(5): > print("Allocating #", i) > out.append(np.zeros(shape, dtype=float).fill(1)) > > > Thanks > Stéfan >

2 1

scikit‑image‑0.10.1.win32‑py2.7.exe 로 모듈을 설치하였는데. dll error 가 뜹니다.
by 박수진 03 Oct '14

03 Oct '14

>>> from skimage import data Traceback (most recent call last): File "<pyshell#4>", line 1, in <module> from skimage import data File "C:\Python27\lib\site-packages\skimage\data\__init__.py", line 11, in <module> from ..io import imread File "C:\Python27\lib\site-packages\skimage\io\__init__.py", line 11, in <module> from ._io import * File "C:\Python27\lib\site-packages\skimage\io\_io.py", line 7, in <module> from skimage.color import rgb2grey File "C:\Python27\lib\site-packages\skimage\color\__init__.py", line 1, in <module> from .colorconv import (convert_colorspace, File "C:\Python27\lib\site-packages\skimage\color\colorconv.py", line 57, in <module> from scipy import linalg File "C:\Python27\lib\site-packages\scipy\linalg\__init__.py", line 159, in <module> from .misc import * File "C:\Python27\lib\site-packages\scipy\linalg\misc.py", line 5, in <module> from . import blas File "C:\Python27\lib\site-packages\scipy\linalg\blas.py", line 145, in <module> from scipy.linalg import _fblas ImportError: DLL load failed: 지정된 모듈을 찾을 수 없습니다. 와 같은 오류가 뜹니다. 무엇이 문제 입니까?

2 1