One more time - FITs question

[Emanuele]

Hi all.

alright, I have read every thread on the subject and tried most of the suggestions in the forum about this.
I have a Master Lum that has been calibrated and registered in Maxim.
Then saved as 32Bit Int.

Opened it up in PixInsight, then rescaled (window popped up asking the range values) to 0-65535, but the result is below (STF)
What am I doing wrong and how do I import this image in PI?

Thank you,
Emanuele

[Philip de Louraille]

Saved as a 32 bit?
2^32  = 4,294,967,296
so your range might be 0 to 4,294,967,295 not 0 - 65,535 (unless you have signed integers and then half these values)

[Emanuele]

Philip, thank you for your help.

I saved the IEEE Float Luminance image in Maxim, in 32Bit Int still in Maxim, and then opened it up in PI but now PI is not asking me to rescale the image.

 
Not sure what to do. How can I force a range in PI?

[Philip de Louraille]

Yes, PI likes to store values between 0 - 1 so it maybe what it is asking. Try entering 0 as the min and 1 as the max.

[Emanuele]

Philip,

what I meant was that PI is not asking me (now) to specify a range for the 32Bit Int Maxim image.

[Philip de Louraille]

hmm.... Open the Format Explorer tab on the far left, and double click on the FITS icons. This will open the FITS Format Preferences window where the Lower and Upper Range are set.
If that is not what you mean/want, then use the PixelMath Process.

[Harry page]

Hi

When I get this I just rescale it using the rescale tool


Harry

[Emanuele]

Yep, the values in the FITS Format Preferences are set to 0 and 1. And also to Ask Always as far rescaling if out of range goes.

How do I set a range in pixelmath?

[Philip de Louraille]

I was looking at the Rescale tool earlier but I was not sure that was the one to use as I do not have an image with that kind of problem to test it on. It does not do anything to my opened images (as they are scaled properly already.)

But it seems easier to use than my PixelMath suggestion!

[Emanuele]

Harry, thanks for the suggestion.

But Rescale is not working for me.    I have the same results as the screenshot above.

[Philip de Louraille]

After the rescale, did you redo an STF automatic adjustment?

[Emanuele]

Yes, Philip, I did. No change from before.  

Here's something strange (see the attached screenshots)
The first screenshot is the the range values of the Luminance and Green masters when importing them into PI.
The second screenshot is the range of values of the Red and Blue masters.

Notice the difference in the range...
well, the Red and Blue masters are rescaled correctly, just by letting PI rescale the image to 0-1.
But instead the Luminance and Green masters are not scaled correctly by PI.

Why do even the Lum, Green and Blue, Red masters have different range values, coming out of Maxim?

[Philip de Louraille]

Can you make your image available? If you do not have access to a public server, place it there
https://public.me.com/pkcyll
(click on the circled white arrow pointing upward to upload your image.)

I have to leave the house in about 1 hr so I may not be abler to look at this with more details until later on this afternoon. (California time.)

[Emanuele]

Hi Philip...

here I am. I placed the images (Lum and Red) in my public dropbox folder.

Hope something can be figured out because it's really good data and it deserves to be post-processed in PI.

E.

EDIT: Since this issue has been solved, I removed the images links for copyright purposes. If anyone wants them again, please ask.

[Andres.Pozo]

Hi Philip...

here I am. I placed the images (Lum and Red) in my public dropbox folder.

http://dl.dropbox.com/u/1427295/NGC2024_Lum.fit

http://dl.dropbox.com/u/1427295/NGC2024_Red.fit

Hope something can be figured out because it's really good data and it deserves to be post-processed in PI.

E.

When Pixinsight tries to open your images it detects a maximum value of 5.79e7. That value is not correct. There are several rows of bad pixels at the bottom result of the stacking process.
When PixInsight shows the "FITS floating point range options" dialog you should set a maximum value of 69000 and the "Truncate all out of range values" option. Doing this the image appears ok in Pixinsight.