[Numpy-discussion] [SciPy-User] Simple pattern recognition
Zachary Pincus
zachary.pincus@yale....
Mon Sep 21 12:57:53 CDT 2009
I believe that pretty generic connected-component finding is already
available with scipy.ndimage.label, as David suggested at the
beginning of the thread...
This function takes a binary array (e.g. zeros where the background
is, non-zero where foreground is) and outputs an array where each
connected component of non-background pixels has a unique non-zero
"label" value.
ndimage.find_objects will then give slices (e.g. bounding boxes) for
each labeled object (or a subset of them as specified). There are also
a ton of statistics you can calculate based on the labeled objects --
look at the entire ndimage.measurements namespace.
Zach
On Sep 21, 2009, at 1:45 PM, Gökhan Sever wrote:
> I asked this question at http://stackoverflow.com/questions/1449139/simple-object-recognition
> and get lots of nice feedback, and finally I have managed to
> implement what I wanted.
>
> What I was looking for is named "connected component labelling or
> analysis" for my "connected component extraction"
>
> I have put the code (lab2.py) and the image (particles.png) under:
> http://code.google.com/p/ccnworks/source/browse/#svn/trunk/AtSc450/
> labs
>
> What do you think of improving that code and adding into scipy's
> ndimage library (like connected_components()) ?
>
> Comments and suggestions are welcome :)
>
>
> On Wed, Sep 16, 2009 at 7:22 PM, Gökhan Sever
> <gokhansever@gmail.com> wrote:
> Hello all,
>
> I want to be able to count predefined simple rectangle shapes on an
> image as shown like in this one: http://img7.imageshack.us/img7/2327/particles.png
>
> Which is in my case to count all the blue pixels (they are ice-snow
> flake shadows in reality) in one of the column.
>
> What is the way to automate this task, which library or technique
> should I study to tackle it.
>
> Thanks.
>
> --
> Gökhan
>
>
>
> --
> Gökhan
> _______________________________________________
> SciPy-User mailing list
> SciPy-User@scipy.org
> http://mail.scipy.org/mailman/listinfo/scipy-user
More information about the NumPy-Discussion
mailing list