Too many stars - looking for tips

[69boss302]

Seems odd saying I have too many stars in my pictures on an astrophotography board!  But when I do some Nebulas...actually almost anything... I get predominant stars where I want the nebula to be the main attraction. Example below...  I do all my calibration frames, stack in DSS, and process in PI. Pretty basic processing.  The image below, Soul Nebula,  was about 57 lights at 300 sec each.  Is the secret more lights but shorter duration?  I really dont love the outcome of a lot of the stuff I do.  As always, any tips much appreciated.

Bob

 

[scotty38]

If you're using PI then use StarNet to remove the stars totally, work on what remains then add (a % to taste) of the stars back using PixelMath.

Edited December 14, 2021 by scotty38

[69boss302]

I have removed stars with Starnet, but never added them back again with pixelmath! Ill definitely try that, thank you!

[vlaiv]

Problem is not with how many stars you have in the image - there is simply that much stars there - take any Soul image out there and compare - you will see same stars in both images.

It is star "radius" that is the problem - you get impression that there are more stars because stars are wider in your image.

There are several things that make stars larger in the image:

1. aperture of the telescope used - smaller telescopes produce larger stars

2. Seeing - poor seeing will produce larger stars

3. Poor mount tracking / poor guiding - again produces larger stars

4. Type of optics - some types of telescopes produce fatter stars - for example large central obstruction causes stars to be slightly more bloated, but large obstruction usually comes with larger aperture - so it kind of evens out. Another culprit is refractor telescope with less than perfect color correction. Fast doublets tend to create fatter stars because of this. You can use special luminance filters like Astronomik L3 filter to combat this.

5. Filters used - some filters increase stars (similar to poor optics - they make stars fatter)

6. Way you process your data.

What you can do about it?

Take care of point 2 and 3 - shoot when seeing is good and try to optimize guiding on your mount. Observe total RMS and try to make it smaller. Don't go by - guiding is fine enough for small scope - want tighter stars? make guiding better

See if using L3 from Astronomik will help with star sizes if you have refractor (and judging by FOV and your signature you are using 73mm ED doublet so it might be worth doing).

See what type of filters you are currently using and maybe try test image without them in order to see if there is any difference. Some LP suppression filters have been reported to cause slight star bloat

Reprocess your data. Sometimes way you stretch can have impact on star sizes. Look into making starless image and processing stars separately to make them smaller.

[scotty38]

Just remember to check the starmask option:

[69boss302]

@scotty38 On a side note, when I remove stars using Starnet++, its a really blotchy image. I was hoping to reduce the stars in the original image and not as much in post...

 

[69boss302]

@vlaivthank you for that great response, much appreciated. I am indeed using a Z73, however I am using no filters, maybe I should try the one you suggest.  I think I have guiding pretty tight, but can see if I can improve it a tad. Im thinking it is the way I process... the individual .fit files dont have bloated stars, so it must be something Im doing in the process...

 

[mike1485]

Some great advice as usual from vlaiv.  

I am not a great fan of complete removal of stars using Starnet++ - as you have observed it tends to leave plenty of blotchy patches.  To some extent these can become less noticeable when the stars are replaced. One little trick using Starnet++ is to apply an inverted Luminance mask to your image and then apply Starnet++.  This has the effect of reducing your stars, making them a little less prominent in the image.

 In your image it looks like many of the brighter stars have become bloated and this may well have resulted when stretching the image.  If you have been using Histogram Transformation for your stretching this process can be heavy on stars - ie, to get the result you want in the nebulosity you end up over-stretching the stars.  I see you have posted in my recent posting on the GeneralisedHyperbolicStretch script, this script can help achieve a better stretch that will protect those stars.

Edited December 14, 2021 by mike1485

[69boss302]

@mike1485 Thank you, and  I think you nailed it, I do use Histogram Transfomation!  And you are also correct in that I comment on your new script and I look forward to giving it a try. My plan is to read about it tonight when I get out of work. Looking forward to it!

 

 

[Budgie1]

Once you've removed the stars with StarNet in PI, you can work on the stars separately.

For reducing the stars I use Morphological Transformation set as below. The "Amount" setting I use gives a 30% reduction, but you can alter it to your own requirements, sometimes I run it twice at 30% to get what I want.

Like with most things in PI, there are a few ways to achieve something, this just the I use most.

[69boss302]

@Budgie1thank you for this. Just shows I have so much to learn still. Thats not a bad thing.... Ill get there. You guys are the best, thank you.

[scotty38]

8 hours ago, 69boss302 said:

@scotty38 On a side note, when I remove stars using Starnet++, its a really blotchy image. I was hoping to reduce the stars in the original image and not as much in post...

 

Bit late getting back to you but all sorted now anyway and a good shout re MT too.

[vlaiv]

11 hours ago, 69boss302 said:

@vlaivthank you for that great response, much appreciated. I am indeed using a Z73, however I am using no filters, maybe I should try the one you suggest.  I think I have guiding pretty tight, but can see if I can improve it a tad. Im thinking it is the way I process... the individual .fit files dont have bloated stars, so it must be something Im doing in the process...

 

What is your guide RMS in arc seconds?

As far as processing goes - there is easy way to check - post your stacked fits in 32bit float without any processing and see what others manage with the data.

If most people come up with fatter stars - then it is due to data (some will use morphological star tightening - which I'm against), but if most manage to get tight stars - then it is down to way you process.

[ollypenrice]

14 hours ago, vlaiv said:

Problem is not with how many stars you have in the image - there is simply that much stars there - take any Soul image out there and compare - you will see same stars in both images.

It is star "radius" that is the problem - you get impression that there are more stars because stars are wider in your image.

There are several things that make stars larger in the image:

1. aperture of the telescope used - smaller telescopes produce larger stars

2. Seeing - poor seeing will produce larger stars

3. Poor mount tracking / poor guiding - again produces larger stars

4. Type of optics - some types of telescopes produce fatter stars - for example large central obstruction causes stars to be slightly more bloated, but large obstruction usually comes with larger aperture - so it kind of evens out. Another culprit is refractor telescope with less than perfect color correction. Fast doublets tend to create fatter stars because of this. You can use special luminance filters like Astronomik L3 filter to combat this.

5. Filters used - some filters increase stars (similar to poor optics - they make stars fatter)

6. Way you process your data.

What you can do about it?

Take care of point 2 and 3 - shoot when seeing is good and try to optimize guiding on your mount. Observe total RMS and try to make it smaller. Don't go by - guiding is fine enough for small scope - want tighter stars? make guiding better

See if using L3 from Astronomik will help with star sizes if you have refractor (and judging by FOV and your signature you are using 73mm ED doublet so it might be worth doing).

See what type of filters you are currently using and maybe try test image without them in order to see if there is any difference. Some LP suppression filters have been reported to cause slight star bloat

Reprocess your data. Sometimes way you stretch can have impact on star sizes. Look into making starless image and processing stars separately to make them smaller.

I'd add a very important '7' to this list. Focus. I'd be very surprised if those stars could not be made smaller by better focus.

Olly

Edited December 15, 2021 by ollypenrice

[vlaiv]

10 minutes ago, ollypenrice said:

I'd add a very important '7' to this list. Focus. I'd be very surprised if those stars could not be made smaller by better focus.

Olly

Good point - best focus is essential!

[69boss302]

@vlaiv Thank you, that is a great suggestion, I have attached my stacked fits to this, maybe someone can make something out of it!   As for focus, I have the ZWO EAF and do a focus routine before PA each time, as well as a temp change and Meridian flip, so hopefully that is enough.  I also have no filter but it seems I should.... 

Rosette.TIF

[vlaiv]

4 minutes ago, 69boss302 said:

I also have no filter but it seems I should.... 

ASI2600mc-pro has integrated IR cut filter, so that should not be a problem - only if doublet scope is causing some bloating?

I'll have a look at attached TIF to see if I can see something useful.

[vlaiv]

Ok, so there is definitively a bit to do with processing.

Take for example this processing - it is only stretched using levels in Gimp:

Stars are a bit larger than they should be - possibly due to seeing, but can also be due to guiding precision:

This is at 100% zoom - two things are obvious. Stars are not pin point - which is strange, given short focal length scope used (again poor seeing and/or poor guiding are to blame) - but also notice scatter around bright stars. That can be due to optics - like a bit of haze / fogging on telescope. Do you have dew shields?

Another possibility is simply haze in the air - if transparency is poor - there will be this sort of halo around bright stars.

If you are not careful - it is very easy to turn that halo into "full star profile". Look at following stretch of the same image - again, I'll be using only levels in Gimp:

See what happened - that halo joined with star core to make very bloated star.

Then whole image looks more "crowded" in the end because of this:

[69boss302]

@vlaiv Thank you! Wow some really good info there. I think I should start with my basics of good guiding.  I do the pa then it does its guiding with almost no input from me, so I just let it go and never pay much attention to it other than look at the graph on occasion.   I do have a dew heater and use it! Its cold up here in CT this time of year.  Ill also check my objective lens for cleanliness but Im fairly fanatical about lens cap etc.  I dont have the guiding specs for this rosette picture, however I did just happen to capture a screen shot from when I shot horsehead, and I think my horsehead came out very nice (for a beginner)...!

[vlaiv]

@69boss302

I took green channel and did FWHM measurement on some of fainter stars and got values of around 4.8px or 8.64" if your sampling rate is 1.8"/px (430mm FL and 3.76µm pixel size).

That is way too large FWHM. Theory says that 70mm scope in 2" seeing and 1" RMS guiding (according to your image above) will have around 3.5" FWHM - so about 2.5 less than what you have.

Either that guiding report is not accurate (maybe wrong FL or pixel size for guider?) or seeing is particularly poor. With 1" RMS guiding and 70mm scope - you need like 8" FWHM seeing to get stars that bloated - like worst seeing ever

 

 

[ONIKKINEN]

There could be frost developing on surfaces other than the objective lens, even inside the tube if you have moisture in there for some reason. I find a thin layer of frost develops quickly and is difficult to see by eye but the camera will pick this up as a dimming/bloating effect.

How do you handle the gear after shooting? Do you open all the caps for several hours to let moisture out before putting the gear back to storage? If you dont, you could have moisture lingering inside the tube which sets as frost to a corrector/filter/inside of the lens or whatever and wkuld be difficult to spot with the objective lens being dry to the outside.

[69boss302]

@vlaiv If I am understanding you correctly, I likely have the wrong camera for my scope?

[69boss302]

1 hour ago, ONIKKINEN said:

There could be frost developing on surfaces other than the objective lens, even inside the tube if you have moisture in there for some reason. I find a thin layer of frost develops quickly and is difficult to see by eye but the camera will pick this up as a dimming/bloating effect.

How do you handle the gear after shooting? Do you open all the caps for several hours to let moisture out before putting the gear back to storage? If you dont, you could have moisture lingering inside the tube which sets as frost to a corrector/filter/inside of the lens or whatever and wkuld be difficult to spot with the objective lens being dry to the outside.

Generally when I bring it back inside, I put it on the sofa and go to bed and worry about it next morning, so it does have a few hours of getting back to room temp.

 

[vlaiv]

34 minutes ago, 69boss302 said:

@vlaiv If I am understanding you correctly, I likely have the wrong camera for my scope?

No, I think you have pretty good match between camera and scope.

1.8"/px is a bit on high side for 70mm scope - using FF/FR might be better option - to get to 2"/px or slightly above - but that is nitpicking compared with issue you now have.

Your scope + normal seeing + your mount and guiding that you've shown in that screen shot should match your resolution. Everything should match and you should be getting around 3.5" FWHM stars (which correspond to ~2.18"/px - and you have 1.8"/px - so slightly off but still very near perfect sampling rate).

Problem is that stars in your image have FWHM of 8.6" and that is much more then expected. This means that stars are not bloated in your images just because of processing - there is something making them very large. About twice as "fat" as they should be given specs of your system.

- Either you were imaging in extremely poor seeing conditions. Maybe your local seeing is very poor? Do you image over rooftops of houses that are heated so there is great differential in temperature? Maybe there is large body of water in direction where you image?

- Another explanation would be that your guiding is not as good as you think it is. According to that screen show - your guiding is ok. It's not great - you could possibly go down to about 0.6-0.7" RMS with your mount, but 1" RMS is not that bad at all. In fact - it is in line with your setup and working resolution. Question is - is reported RMS accurate or not? It can be off if you entered wrong values for either pixel size or focal length of guide scope.

- Third and unlikely possibility is that something in optical train is blowing up your stars. Maybe that flattener, or perhaps fact that you are using ED doublet - but I'm not seeing too much of blue fringing in your images - so I don't really think that is the case.

In fact - maybe it is flattener? Looking at this spot diagram:

RMS radius goes from ~7.2 up to almost 10µm - so that is two pixels up to 3 pixels for radius.

This equates to 7.2" RMS diameter up to ~10" RMS diameter - with flattener. That is significant bloating of the PSF.

Telescope without flattener should have 3.52" diameter of airy disk - so twice as small.

If this bothers you - maybe try imaging without flattener once and see if stars in the center of the frame will be better (of course - you'll have distortion at the edges due to field curvature). Maybe get different FF/FR for your scope if this is causing issues for you?

 

[69boss302]

 

Wow so much good information, thank you again!   OK good news on the camera / OTA match, I did do some homework before buying it all and thats what I came up with so Im glad to hear that.  I am actually in a pretty good spot for viewing, top of a hill, no water and Bortel 3/4. Not near any big cities. It gets pretty dark out here.  I will read up on getting better tracking, I know thats vital, thank you for that.  As for visual conditions, when I start out its not so bad, but admittedly I do go in and get some sleep.  Maybe the conditions change throughout the night.  I do know that a few nights ago wasnt ideal, it had a thin layer.  Ill pay more attention to that next time.  Now... on to the field flattener.... Thats a good possibility. I remember thinking how sharp my center stars were before I got the FF (stacked tiff image attached).  I decided to stay with the one 'made' for my Z73 and my assumptions were that it would be best for it but maybe Im wrong on that. Have you any recommendations for a better quality one?  THank you for helping me nail this down!

Bob

 

Andromeda2.tif