PixInsight - newbie on trial licence

[julian489289]

Hi

I am trialing PixInsight and trying to process my Leo triplet subs.  I have managed to pre process all of my subs approx 20 lights, darks, flats & bias and arrived at stacked & aligned image, however the resulting image has gradients/vignetting??

When I try to use DBE, I get the following.  I have tried all sorts of adjustments but can get no better

What can I do to correct this before I move onto  background neutralisation & colour calibration etc....

I hope  someone can help as this is putting me off of purchasing the full software.

Thank you

[Scooot]

I'm no expert but after calibrating with flats, darks etc it shouldn't look like that. Did the flats look ok? Maybe you could post a pic of the auto stretched master flat.

[RSM]

Hi, I usually end up with a very uniform, but heavily coloured background after stacking which both ABE or DBE take care of quite easily. My flats take out the vignetting which looks to be appearing on your stacked image. If you stretch an individual flat do you get a similar pattern or is it uniform. 

[julian489289]

Thanks guys

Here is the master flat stretched

Doesn't look right  does it??

 

[Scooot]

That looks ok to me. So I think something's gone wrong when you've calibrated and integrated the lights.

check what you've done against this, http://www.lightvortexastronomy.com/tutorial-pre-processing-calibrating-and-stacking-images-in-pixinsight.html

[wimvb]

Your flats overcorrected the image. Taking somewhat darker flats can help. For this image, you can recalibrate and stack without flats and/or use dbe:

1. Place samples manually some in each corner, along the edges and towards the middle. A total of up to 15 samples can be enough

2. Increase sample size, larger samples that do not touch or cover stars (up to 15 pixels or so)

3. Increase tolerance if necessary

4. Use division as correction method

5. Try with and without normalisation

6. Don't discard the extracted background. Stretch it with STF and examine it. Gradients must be gradual. Blotchy, multicoloured background models indicate that the dbe process didn't work as it should.

These steps should remove the vignetting

Good luck

[julian489289]

Thanks Guys

I try to get 2/3 histogram for my flats, so maybe I will reduce to 50%?  Should I do this?

I will restack & recalibrate without flats and see what the result is.

Thanks again, will update the post with the results.

 

[cgarry]

Could you please share one of you lights and one of each of your calibration frames?

Cheers,
Chris

[julian489289]

Light

L_2015-02-21_23-15-42_M66_0018_ISO800_240s__7C.CR2

Dark

D_2015-02-22_00-56-18_M66_0001_ISO800_240s__11C.CR2IMG_6174.CR2IMG_6174.CR2IMG_6174.CR2

Flat

IMG_6174.CR2

Bias

IMG_4369.CR2

 

Chris, I have included one of each above, hope the links work if not I will send a link via dropbox one uploaded.

Regards

[julian489289]

https://www.dropbox.com/s/8rpu6eakdfqiz9u/L_2015-02-21_23-15-42_M66_0018_ISO800_240s__7C.CR2?dl=0

https://www.dropbox.com/s/kc0ewzr6em75npc/D_2015-02-22_00-56-18_M66_0001_ISO800_240s__11C.CR2?dl=0

https://www.dropbox.com/s/dl9f7g3u7y42oie/IMG_6174.CR2?dl=0

https://www.dropbox.com/s/c1onbh2g1mvazk9/IMG_4369.CR2?dl=0

Just incase above did not work Chris.

[cgarry]

Perfect, thanks.  I'll report back when I have had a play!

Chris

[ollypenrice]

You don't have vignetting, you have inverse vignetting, as Wim suggests. This may well be because your flats were not corrected by their own darks. I don't believe flats need, literally, their own darks because they are short exposures for which a master bias will do perfectly well as a dark. But you do need to give your stacking software something usually called a dark flat. Call it what you will, but a master bias makes a perfectly good dark flat. (I would prefer to call it a 'dark for flats' because - well - that's what it is!)

Olly

[Oddsocks]

Julian4892989,

 

Check also in you camera setup menu that image portrait / landscape auto rotation is turned off.

When I open the RAW CR2 in PI the bias is horizontal and the lights, darks, flats are vertical making image calibration difficult since I don't know whether to manually rotate the bias frame clockwise or anticlockwise before using it in the calibration sequence.

William.

[julian489289]

OK, I have reworked all calibration files inc Flats as per Light Vortex Astronomy tutorial as suggested by Richard.

Better than before bu still not right...This is the stacked aligned image

 

This is the DBE background rejection

and this the divided result

I am now to repeat but without stacking the master flat..........

[wimvb]

AFAIK, PI can handle different orientations of the camera,

https://pixinsight.com/forum/index.php?topic=8751.0

@julian489289, try removing the black edges left over from image integration, before doing the DBE process.

[julian489289]

Yes I did that with Dynamic crop before DBE but the screen shot was of the pre crop 

[cgarry]

6 hours ago, julian489289 said:

https://www.dropbox.com/s/8rpu6eakdfqiz9u/L_2015-02-21_23-15-42_M66_0018_ISO800_240s__7C.CR2?dl=0

https://www.dropbox.com/s/kc0ewzr6em75npc/D_2015-02-22_00-56-18_M66_0001_ISO800_240s__11C.CR2?dl=0

https://www.dropbox.com/s/dl9f7g3u7y42oie/IMG_6174.CR2?dl=0

https://www.dropbox.com/s/c1onbh2g1mvazk9/IMG_4369.CR2?dl=0

Just incase above did not work Chris.

This is the image that was generated when I calibrated your image file:
https://dl.dropboxusercontent.com/u/1139742/outgoing/L_2015-02-21_23-15-42_M66_0018_ISO800_240s__7C.tif

Warning, this file is 106,073 KB

To my eyes this calibrated image does not show the over calibration issue that your post showed so I do not think there is anything wrong with you data.

Cheers,
Chris

[julian489289]

Well, I have just repeated the process without stacking any flats.  Same appearance, so I have no idea what I am doing wrong in the calibration and stacking process.

 

I have followed the tutorial and suggested settings , any ideas anyone please??

 

 

 

 

 

 

[wimvb]

If you don't mind sharing the stacked, but otherwise unprocessed xisf image, I can have a look at it. Either publish a link here, or pm me.

Cheers,

[julian489289]

Wim, I have the lights & calibration subs uploaded on drop box I think.  Not sure where i am going wrong in the stacking process though.  If OK when I get back home later, I will drop the link onto here for those files.  If you get time and could stack them in PI to see if you get the same result that would be fab, no worries if not though dude.

 

[julian489289]

Sorry, here is the link I had to reload them all to dropbox

https://www.dropbox.com/sh/4z3tlw8ncpsmdti/AABoVXS-vpraWh2aeEC_BrcFa?dl=0

 

[wornish]

I am in a similar situation to you as a newbie to Pixinsight and I am struggling to get the Image Integration Pre processing workflow to work properly.  I hope you don't mind but I downloaded 10 of your lights, 10 bias frames, but no darks and tried them in Pixinsight.  I followed the tutorial steps on here

http://www.lightvortexastronomy.com/tutorial-pre-processing-calibrating-and-stacking-images-in-pixinsight.html

The background issue was fixed  but the stars were not aligned at all.

I thought I'd try the same raw files in DSS it worked superbly.  The autosave.tif did have some normal vignetting but this was immediately removed with the Automatic Background Removal process  when I loaded it into  Pixinsight .

Your data is amazing even only using 10 frames.

I hope someone on this thread can show us how to use the Pixinsight calibrating and stacking  steps.

 

[wimvb]

Sometimes Pi sees hot pixels as stars. If this is the case, the image offsets will all be 0 px. In the star alignment section, you can increase the noise threshold and wavelet parameters one or two steps. During calibration you can add cosmetic correction as a hot pixel filter. Usually this solves the issue.

[wornish]

Just realised what I did wrong I completely missed out the star alignment process step from the long list of steps in the above link..

So did it again on 10 lights and 10 bias frames only and the resulting Calibrated, DeBayered, Aligned, Stacked file is great.  Just ran a basic Automatic Background Removal process on that and then stretched., plus a few tweaks.

The data is excellent.  So if you follow ALL the steps in the light vortex tutorial the OP issue is fixed.  Learnt a lot and never had such good data to practice on.

 

Hope this helps 

[julian489289]

Dave, prior to looking at PI I did use DSS to stack and tweak in PS CS6.  Got a great result from DSS and not too a bad final image after PS.

However, I thought that I would try PI and hoped to get an even better image result but I am somewhat disappointed and reluctant to spend E280.

Can I just ask why you dropped out the darks?  Was it the darks that were causing the vignetting issue above?