search

leginon-org / leginon-redmine-archive

1 stars 0 forks source link

ace2 on FSU Krios #596

Open leginonbot opened 5 months ago

leginonbot commented 5 months ago

Author Name: Scott Stagg (Scott Stagg) Original Redmine Issue: 596, https://emg.nysbc.org/redmine/issues/596 Original Date: 2010-05-28

I can't get ace2 to work consistently with images from our Krios. The problem, I believe, is that the edge detection fails on the high mag. images that we are taking. The zeros are relatively close to the origin, and ace2 seems to miss them. When this happens, it misestimates the astig and doesn't elliptically average correctly. I think the best fix for this would be to make an option where you can turn off the edge detection and just circularly average. I don't know who to make a watcher on this, so if I have put the wrong people, please drop yourselves and add the correct ones.

leginonbot commented 5 months ago

Original Redmine Comment Author Name: Neil Voss (@vosslab) Original Date: 2010-06-16T23:04:56Z

If edge detection fails it uses circular average. As a temporary workaround, make the edge threshold really high like 0.5 instead of 0.001 and it should work like you want.

leginonbot commented 5 months ago

Original Redmine Comment Author Name: Scott Stagg (Scott Stagg) Original Date: 2010-06-17T10:58:40Z

That definitely helped, but it still fails on a large percentage of images. What percentage of images does it succeed on at Scripps?

leginonbot commented 5 months ago

Original Redmine Comment Author Name: Neil Voss (@vosslab) Original Date: 2010-06-17T12:21:09Z

It usually works on 100% of any image you would expect it to work, but it was really designed around Tecnai 2. I hope to dive into the code, fix it so it works on more microscopes and then eventually publish it.

leginonbot commented 5 months ago

Original Redmine Comment Author Name: Anchi Cheng (@anchi2c) Original Date: 2010-06-17T13:38:33Z

Try use Binning option in runAce2.php

I put some behavior description in the pop-up help. In your case, by using a large binning, you zoom into the center of the power spectrum so that ace2 can find the modulation in the range it can handle.

My data are normally at 100kx on Tecnai2 so I have similar problem.