Project

General

Profile

Forums » Leginon »

Frames not saved

Added by Bob Grassucci over 11 years ago

We have a new error. Leginon is not saving frames even when we have save frames selected on the GUI. See JPG of launcher output window

DM output below. The gatan-k2.py files is eddited as it was before except for the new aditions you made and sent.
An error occurred acquiring an image for SerialEM:
Failed to open above file: No such file or directory

An error occurred acquiring an image for SerialEM:
V:Image\20130320_15333000\frame_001.mrc

frames_error.jpg (239 KB) frames_error.jpg

Replies (34)

RE: Frames not saved - Added by Bob Grassucci over 11 years ago

Is frame exposure time still hard coded? dmsem.py seems to have a different text for the self.dosefrac_frame_time than gatank2.py did. How do I set the frame time? There is no place to do it on our GUI.

RE: Frames not saved - Added by Jim Pulokas over 11 years ago

Bob,
I am wondering if you are not completely updated, or the old version is somehow getting used. Verify the svn revision in the Help->about menu. I am attaching a screen shot of what the camera settings GUI should look like (in Manual Acquisition settings, for example). There should be a box for "exposure time per frame".
Jim

camera-settings.png (64.1 KB) camera-settings.png

RE: Frames not saved - Added by Anchi Cheng over 11 years ago

HI, Bob,

The schema update problem is probably caused by you trying to do this on Windows. Try in on the LInux box. I think the syntax for looking for the version from svn checkout I wrote does not work on Windows. Will have to debug that in the future.

RE: Frames not saved - Added by Bob Grassucci over 11 years ago

Hi All,
Running the update on the linux box solved the problems. Thanks.

RE: Frames not saved - Added by Bob Grassucci over 11 years ago

Now that we have the frames we have a few questions:

1 What is displayed in the user interface when collecting frames? 1 frame or the whole projected z stack?
2 What type of mrc stack format is saved? Floating point or integer? When I first tried loading it back into DM I got a black image (all zeros). We were able to get something later with an other viewer but want to be sure of exactly what we need to look for.
3 Are these frames hardware corrected or not corrected at all? We are seeing some strange horizontal lines at the bottom of the image.

Thanks.

RE: Frames not saved - Added by Anchi Cheng over 11 years ago

1. whole projected z stack

2. unsigned 16-bit integer, as in Serial EM. (MRC mode 6)

3. The frames should be hardware corrected because DM must do so before counting. Please attach an image or screen shot of the strange image. The projected z stack sent back to Leginon should look exactly like what you would get in DM gui using DM counted or super-resolution mode in dose-fractionation mode.

RE: Frames not saved - Added by Anchi Cheng over 11 years ago

By the way, yes, I've noticed that DM does not read unsigned 16-bit integer MRC.

RE: Frames not saved - Added by Bob Grassucci over 11 years ago

3. Please find attached one frame as viewed in a viewer designed by Rob L. in our lab. Please note the horizontal stripes in the frame. This is why I questioned if it was gain corrected or not. I also noticed that the images displayed in the GUI are much lower in contrast than what is displayed by DM.

example_mic.png (1.42 MB) example_mic.png

RE: Frames not saved - Added by Anchi Cheng over 11 years ago

That image looks like electronically out-of-sync. We've gotten it once and system off/on restart in DM fixed it. Do you get the same artifact in the frames in DM in parallel ?

Yes, Leginon image viewer does not have the best contrast or compression algorithm when zoomed out. You may want to check the files in an indepenedent viewer like the one Rob made to confirm that the data is o.k. With all that hardware on gatan computer, the graphic is certainly great.

(26-34/34)