Going from Maxim 5.23 to PI

[rwstanley]

I am trying to go from calibrating and stacking my images in Maxim 5.23 to PI 1.8.  Images are captured in Maxim.  The problem I am having is that when calibrated in PI, I am getting black spots in some of the bright stars where it is saturated.  Is there anything I can do in PI to eliminate this besides using the clone tool?

Thanks,
Ron Stanley

[sreilly]

Sounds like a bad dark master maybe? Do you have optimize checked for you dark frame?

-Steve

[rwstanley]

I created the masters in PI, and yes I have optimized checked for dark frame.  I have noticed that any images I have calibrated in Maxim when opened in PI show the same spots, however when I stack in Maxim and open the stacked images in PI there are no black spots.

Thanks,
Ron

[Josh Lake]

Have you tried unchecking the Clip High Range in the Rejection section of Image Integration?

[rwstanley]

I did try that however the black spots have a value of 0.

-Ron

[Carlos Milovic]

Do all the saturated stars have a cero value in the core?

Just as a workaround, if that is the case, you may try PixelMath to change all the zeros to ones (with something like "iif($T==0.0,1.0,$T)" )

[Juan Conejero]

I created the masters in PI, and yes I have optimized checked for dark frame.  I have noticed that any images I have calibrated in Maxim when opened in PI show the same spots, however when I stack in Maxim and open the stacked images in PI there are no black spots.

This problem has nothing to do with dark frame optimization. This looks like a FITS compatibility issue. FITS raw frames loaded in the ImageCalibration tool or the BatchPreprocessing script must be in 16-bit unsigned integer format. Are you using raw frames stored in a floating point or signed integer format, by chance?

By the way, these interoperability problems (among many more FITS deficiencies) are impossible with the new XISF format. Hopefully it will be supported by image acquisition applications in the future.

Have you tried unchecking the Clip High Range in the Rejection section of Image Integration?

This is another possible explanation. If star cores are saturated in all of the images, this option will reject them and yield zero. Anyway, Clip high range is disabled by default in ImageIntegration.

As Carlos has suggested, PixelMath can be used to fix these artifacts after the fact.

[rwstanley]

I think this has something to do with running Debloomer in Maxim.  I need to use it because I use a QSI 532 camera.  When I run it it sets the BZERO value from 32780 to 0.  I'm not sure what this means but I think it has to do with my black spots.  In any case I used the expression which Carlos provided in Pixel Math and it removes the black spots from my calibrated images so that is how I will handle it in the future. 

Thanks much for your help!
Ron

[jkmorse]

Ron,

I used to have this problem all the time and came up with a number of solutions that I have discussed re this and other FITS compatibility issues a number of times here, particularly between MaximDL and PI.  There is any easy fix for the problem you are experiencing.  Just go into MaximDL and do a batch convert of the files from 16Bit to 32bit floating point.  That should solve this particular problem, but if you still have issues in ImageCalibration, insert the following in the "Input Hints" section: lower-range 0 upper-range 65535

That should take care of the black holes.  If you are having any other weird effects, such as images appearing blown out, just let me know and I will copy here my entire fix note from an earlier set of posts. 

Best,

Jim

[jkmorse]

Ron,

In case you are interested, here is a link to one of my earlier posts on this issue in more detail, though you need to read down to get to my 16Bit to 32Bit fix for black star centers:

http://pixinsight.com/forum/index.php?topic=8021.msg52728#msg52728

Best,

Jim

[rwstanley]

HI Jim,

Thanks for the information.  I did a search for the problem in the archives but I missed your thread.  I will give your suggestions a try.

Thanks again,
Ron

[rwstanley]

Hi Jim,

Converting to IEEE Floating during calibration seemed to do the trick. 

Thanks,
Ron