Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

M81, M82 and the IFN

swag72

By swag72
in Imaging - Deep Sky

Recommended Posts

  • Replies 48
  • Created
  • Last Reply
swag72

swag72

Posted
  • Author
Posted
8 hours ago, harry page said:

Hi

did you have a go with the pixel math process Icon :)

 

Harry

I did *shock* :)  Honestly though, I wasn't keen on what it did in this instance...... but a good tool to have for sure :) I do appreciate you showing it to me Harry ??

Link to comment
Share on other sites
AndyUK

AndyUK

Posted
Posted

Damn, this is a good image, Sara!  And it's definitely worth looking at the higher res version... :). 

(Even if I could capture such data, I'd have no idea at all how to start processing something like this - I'm sure if I tried M81/82 would end up as burnt out blobs... and I'd probably end up cropping them out!)

Link to comment
Share on other sites
swag72

swag72

Posted
  • Author
Posted
On 03/08/2018 at 08:06, Saganite said:

Glorious Image !

I was looking at at them both last night but they didn't look like that.  :wink: :grin:

Do you know I think that imagers have a lot to answer for in that regard LOL!!! ........ starting out people think they will see what we produce after many hours of acquisition and processing :( 

Link to comment
Share on other sites
swag72

swag72

Posted
  • Author
Posted
On 04/08/2018 at 17:14, davew said:

Elysian !  It's not often you can use that word eh ?

I applaud your dogged determination and skill. Thank you,

Dave.

I had to look that up! Thank you for teaching me a new word!!

Link to comment
Share on other sites
Barry-Wilson

Barry-Wilson

Posted
Posted

Well done Sara on a fine image. These IFN images are difficult to capture and to process leaving a semblance of a ‘natural’ brightness balance - you have certainly done that. Congratulations ?.

Re the PI process - Harry’s link is primarily used in NB HST images and does not result in larger stars. So not sure how it will work on your broad band image.

Another ‘trick’ if you want to remove some magenta in stars or star halos, is to invert the image (Image>Invert), then use a percentage of SCNR Green (set a percentage of your choice), then invert once again. This works because the magenta when inverted presents itself as green. You may find though that using this technique on your image may remove some magenta you wish to keep: you’ll have to experiment and use your PS skills to layer in/out the areas to discard or keep etc.

HTH.

Barry

 

Link to comment
Share on other sites
swag72

swag72

Posted
  • Author
Posted
13 hours ago, Barry-Wilson said:

Re the PI process - Harry’s link is primarily used in NB HST images and does not result in larger stars...........

I have to disagree with this Barry...... I have attached below a comparison of the two images (small crop of a magenta star) and when you flick between the two images the PI Magenta reduced star has bigger stars. Then I saved each one and loaded them into PI and took a look at the Sub Frame selector script...... eccentricity, FWHM and noise were ALL higher in the PI processed 'reduce magenta stars'......... if FWHM is bigger then that confirms to me that the stars are indeed bigger (which is what I see) 

This is a narrowband image ..... so evidence by eyeball says to me that the stars are bigger when this process in PI has been used and then this is further confirmed by PI measurements of FWHM. I wanted to put it through the PI sub frame selector as there was a possibility that the stars only looked larger because they are whiter overall....... but I can't argue with the hard evidence.

So an interesting observation I think and one that is not an ideal compromise.

Compare stars.jpg

Link to comment
Share on other sites
Barry-Wilson

Barry-Wilson

Posted
Posted

From your description I thought this image was a mixture of Ha and LRGB as oppose to SHO in which the PixelMath magenta stars reduction tool first arose. As I mentioned I have never used it on a broad band mixed image, unless I have not read the thread correctly (quite possible as am on holiday ?) only ever on an HST image.

Link to comment
Share on other sites
swag72

swag72

Posted
  • Author
Posted

Sorry Barry I've probably not been clear...... I didn't use it on this LRGB image. But when I did try it on a NB (SHO) image I was surprised to see the stars get bigger..... as shown in my side by side crop of SHO data in my previous post...... definitely made stars bigger.

I posted the comparison here as that's where Harry bought up the process.... but the data I've used in the comparison is NOT from the original data in the OP.

Link to comment
Share on other sites
sloz1664

sloz1664

Posted
Posted

Back from holiday and just picked this up. 

Stunning image Sara. You're perseverance certainly pays off. You may have spent 64 hours imaging the subjects, but how many hours of dedicated processing did it take to render this fine image.

Steve

Link to comment
Share on other sites
Barry-Wilson

Barry-Wilson

Posted
Posted

Ok, I see.

The PI formula swaps colouring depending on the value in a pixel. Just thinking aloud (ie guessing) it may be that by changing the colour in a pixel the PI FWHM tool now measures the value in the star halo whereas before it did not. 

I also have no idea whether using this PixelMath on a .fit or .tiff image makes a difference (not sure what your image format is).

Have you tried the image inversion trick - does that give you a better result?

Link to comment
Share on other sites
harry page

harry page

Posted
Posted
On ‎06‎/‎08‎/‎2018 at 10:40, swag72 said:

@harry page - Here's a question Harry...... whenever I use that process that you posted, it makes the stars significantly larger than they were before.... Of course I don't want large stars!! Is this normal for this process?

Hi

I have not noticed any increased star size , perhaps they appear larger  when the magenta is removed with a lighter colour

I will go investigate :)

 

Harry

Link to comment
Share on other sites
swag72

swag72

Posted
  • Author
Posted
2 hours ago, harry page said:

Hi

I have not noticed any increased star size , perhaps they appear larger  when the magenta is removed with a lighter colour

I will go investigate :)

 

Harry

Would you agree that if I put the 2 crops into the subframe selector, they should still have the same FWHM? And if one now has a FWHM larger than the other that means that whatever process has been carried out has changed the size of the stars?  It could easily be that I misunderstand the subframe selector script :) 

Link to comment
Share on other sites
harry page

harry page

Posted
Posted
1 hour ago, swag72 said:

Would you agree that if I put the 2 crops into the subframe selector, they should still have the same FWHM? And if one now has a FWHM larger than the other that means that whatever process has been carried out has changed the size of the stars?  It could easily be that I misunderstand the subframe selector script :) 

Hi

Looking at this a bit harder maybe there is some brightness increase with using this tool , giving the illusion ( IMO ) that the star is increasing in size , perhaps this is what the software is looking at

I prefer to use the old eyeball and I personally see no increase in size of the stars

 

Harry

Link to comment
Share on other sites
Barry-Wilson

Barry-Wilson

Posted
Posted
2 hours ago, harry page said:

Hi

Looking at this a bit harder maybe there is some brightness increase with using this tool , giving the illusion ( IMO ) that the star is increasing in size , perhaps this is what the software is looking at

I prefer to use the old eyeball and I personally see no increase in size of the stars

 

Harry

This is my thought too.

Would it be correct that if each pixel brightness value increases as a result of the PixelMath process, that the overall maximum brightness would increase, including the spread of brightness across the star profile (but not the overall width of the profile as the process is not creating new brightness values in pixels adjacent to the star, ie just outside of the halo)?

If so, would the measure of FWHM also increase but not the overall star width?  In my mind I’m picturing the star profile (just like you see in PHD2) with the width of the base of the distribution being the same but the bell shaped curve being wider half-way up the curve.

Anyway, regardless of what is happening at the pixel value, I’m not sure that Sara prefers the resultant effect.

Sorry this has not ticked the box for you.

Might still be worth trying the other PI fix I mentioned Sara. There are other PI processes I would suggest but I know it is not your chosen weapon of choice.

Good luck with a solution ?.

Link to comment
Share on other sites
swag72

swag72

Posted
  • Author
Posted
8 minutes ago, Barry-Wilson said:

 

Might still be worth trying the other PI fix I mentioned Sara. There are other PI processes I would suggest but I know it is not your chosen weapon of choice.

 

Worked a treat thank you Barry ????

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.