search

imagej / imagej2

Open scientific N-dimensional image processing :microscope: :sparkler:
https://imagej.net/
BSD 2-Clause "Simplified" License
1.2k stars 335 forks source link

Pluggin segmenting less number of particles using ImageJ version 1.53f #271

Open TeresaSilva opened 3 years ago

TeresaSilva commented 3 years ago

I am trying to use the same pluggin/method that was used in an old version of imageJ and compare it to the current ImageJ version. So for comparison purposes, I extracted organisms from the same picture, and the segmentation of images using the same pluggin produces much fewer silhouettes (<25%) with the recent version of ImageJ (1.53f) than when using an older version of ImageJ (1.1). I have tried, but it did not work, to troubleshoot imageJ according to the indication in the website: https://imagej.net/Troubleshooting.html#The_same_plugin_gives_different_results_on_different_machines.21

It seems that the current version of ImageJ Analyze of particles is not working properly and only detecting particles with a smaller area 0.049-5.64mm. Even when I select the option of 0-infinity in the current ImageJ version, I obtain much smaller size organisms (0.00010 -5.69mm) for the same image. While using an old version of ImageJ (1.1) for the same image it was detecting particles from 0.049 to 20.26mm.

Is there anything I can do to add to the pluggin to make sure that the data are replicated, i.e. the extraction of particles is constant and does not vary with the ImageJ version? Or is there something else that could be affecting this? Thanks in advance.

image The image attached shows the position of the animals detected using the old version- of imageJ 1.1 (black colour) and new version 1.53f- (red colour). Y-axis indicates the position of particle detected over the y axis in the digitized image and X-axis indicates the position of the particle detected in the x-axis in the digitized image.

imagejan commented 3 years ago

This might be the same issue as discussed here on the forum. If yes, it is fixed in the latest daily build of ImageJ1 (ij.jar).

TeresaSilva commented 3 years ago

Thanks, Jan, It tried to remove the exclude and it did not work. And I even installed the most recent version of ImageJ 1.53h9. It is rather strange that in the new version we get much less particles.