Hi Iman,
One thing you should do is get rid of the alternating brightness of the lines.
Process>Filters>Convolve with the following kernel (normalized) will work:
1
2
1
If necessary, you might then try some edge-preserving blur to reduce the remaining noise, e.g. the Thresholded Blur (radius about 4, rather large threshold, e.g. 30):
http://imagejdocu.tudor.lu/doku.php?id=plugin:filter:thresholded_blur:startThen it should be easy to threshold the white part of the bubbles. For the thresholding the black areas of the bubbles you have to get rid of the uneven background; this is clearly more difficult and probably needs some manual work. You might also try the Trainable Weka Segmentation of Fiji for the black areas.
Michael
________________________________________________________________
On Feb 25, 2015, at 14:48, Iman01 wrote:
> Hello every one,
>
> I'm trying to measure bubble diameters inside a reactor. The original images
> were not clear or transparent enough, so I had a friend to work on them in
> photoshop to get more clear bubbles inside the image. Still I cant process
> the image in imagej. I cant set the threshold to match the B/W bubbles
> inside the new picture. Here are the original and the edited images. So is
> there any way to improve the quality of this image to be processable?
>
>
> <
http://imagej.1557.x6.nabble.com/file/n5011763/bbb.jpg>
> <
http://imagej.1557.x6.nabble.com/file/n5011763/233322.jpg>
>
> Thanks for your help in advance
--
ImageJ mailing list:
http://imagej.nih.gov/ij/list.html
Locked
