Hmm, I don't see the problem.
From the examples the myelin is easily to be detected (
http://picasaweb.google.com/lh/photo/-14v2jvU9p580_6orU8AYg
) and with some logic the axons are more or less good to be isolated
(say: evaluate all objects inside the myelin (background or filled
myelin), group these objects by mean intensity, delete the darker ones
(myelin inclusions), delete the ones with large perimeter/area or p2a
(inter-fibre objects)), the rest possibly after some cleaning will
hopefully be the axons, nicely isolated by the closed myelin.
However, (i) the images shown are not just the base for analysis,
better use jpg or tif (in the gif image visible brightness is not
related with pixel values!), (ii) the mosaic is not well performed and
(iii) IMHO the staining could be considerably improved (compared to
nerv section images on the web). For reasonable quantitative image
analysis some diligence is necessary!
Regards
Karsten
Am 19.11.2008 um 00:50 schrieb Va Paula Sicurello:
> Hello Everybody,
>
> I have a user here who needs to be able to distinguish between two
> different sizes of axons, those 50um and under and those 51um and
> over. I think she measures along the longest side because some are
> more oval or odd shaped than round.
>
> The images are at
>
>
http://www.vmrf.org/researchcenters/confocal/gallery.html>
> The images of interest are on the bottom of the page with no
> description, they are .gif format.
>
> Because these sections are stained with toluidine blue and saved as
> gray scale, Image Pro Plus is having a hard time distinguishing
> between the two sizes. It sees the image as one big gray image with
> no distinct internal features. We haven't tried ImageJ as yet.
>
> I know the images are of poor quality, out of focus and bad
> resolution but this is what we have to work with. I'm guessing that
>
> If you have any suggestions as to how to make this counting process
> easier please let us know your suggestions.
>
> Reply to either
[hidden email] or to me.
>
> Thank you for any help you may send us.
>
> Paula :-)
>
> Paula Sicurello
> VA Medical Center San Diego
> Veterans Medical Research Foundation (VMRF)
> Core Microscope Facility, room B141
> 3350 La Jolla Village Dr., MC151
> San Diego, CA 92161
> 858-552-8585 x2397
>
>
>
>
Locked
