http://imagej.273.s1.nabble.com/Question-about-Mice-Axon-Count-using-ImageJ-tp5000251p5000252.html
> Hello, everyone. This is Steven. Right now I am working on quantification of
> axon in mice model. I am trying to figure out an efficient way to quantify
> my samples. I tried
> 1st:using threshold(black and white), in this step some of the axon detail
> was lost, some area is either too white or black, how could I avoid this, is
> there a way to locally treating the contrast
> 2nd:using nucleus count, this step was not really efficient, I am still
> trying to improve the count as it counts much less than what is out there
> This is basically what I have done, I am willing to listen to any
> suggestions or other methods to quantify. Thanks, and I attached one of the
> image that I took to give an example to what I am working on.
>
http://postimage.org/image/z5jronpy9/> <
http://postimage.org/image/z5jronpy9/>
>
- Trainable Segmentation (it's like a fancy threshold). After