Analyze particles runs slower when launched from a macro

[Pulceblue]

Dear all,
I have noticed that since recently Analyze Particles runs pretty slow for image sequences.
It used to be very fast in previous version of FIJI
Has anyone a clue?
Thanks

[Pulceblue]

I can confirm that for the very same image stack Analyze Particles runs much fasten in in the ImageJ version 1.47v compared to the currently latest version 1.51d.
Why is it running slower in newer versions?
Thanks!

[Rasband, Wayne (NIH/NIMH) [E]]

> On Jun 27, 2016, at 8:17 AM, Pulceblue <[hidden email]> wrote:
>
> I can confirm that for the very same image stack Analyze Particles runs much
> fasten in in the ImageJ version 1.47v compared to the currently latest
> version 1.51d.
> Why is it running slower in newer versions?

It is not normally possible to answer a question like this without having a macro that reproduces problem. The following test macro creates 50 slice stack with ~41,000 particles and runs the particle analyzer on it in about the same time (less than 3 seconds) on both ImageJ 1.47v and 1.51d. The newImage() function on ImageJ 1.47v does not have a “random” option so you will need to save the test stack on 1.51d and open it in 1.47v.

-wayne

  newImage("Untitled", "8-bit random", 1024, 1024, 50);
  run("Gaussian Blur...", "sigma=5 stack");
  setAutoThreshold("Default");
  run("Set Measurements...", "area mean min centroid");
  run("Clear Results");
  t0 = getTime;
  run("Analyze Particles...", "stack");
  print(nResults+" particles, "+(getTime-t0)/1000+" seconds");


> --
> View this message in context: http://imagej.1557.x6.nabble.com/Analyze-particles-runs-slower-when-launched-from-a-macro-tp5016733p5016734.html
> Sent from the ImageJ mailing list archive at Nabble.com.




--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html