Back to forums

Integration fails in Pixnsigt

Acquisition Registration Deep Sky M106 Messier Objects
12 replies•213 views
Patrick Graham avatar
Patrick Graham avatar
I haven't had this happen before but when trying to integrate the 146 subs I shot of  M106, I keep getting the same "not enough stars available to match" error message.  Visually the subs look great and even removed bad ones with the subframe selector tool.  Each sub was shot through a ZWO 2400MC-Pro attached with the L-Enhance filter.  My OTA is the Stellarvue SVX130T. I've tried different settings as instructed by Warren Keller's book, "Inside PixInsight".  Still no success.

Field of view is 3.24 degrees X 2.16 degrees.  Resolution is  1.92" x 1.92"  Above is a sample subframe.

haven't had this issue with any other targets.  Anyone know what I'm doing wrong or suggestions for settings to get this to work?

Pat Graham....Oops....shoulda put this under processing techniques.   My bad
Almos Balasi avatar
Almos Balasi avatar
Are you using the weighted batch preprocessing script? With this script the same problem happened to me. The error is sometime due to the meridian flip, if I rotate the recordings after the flip with 180 degrees, than the integration works (usually for mono images). Sometime is the same problem with CFA images, and it is probably your case,when the detection of bayer matrix selector is set to auto. I change it to RGGB for all light frames (you can set  it in Control Panel -> CFA images) and it works again. By the way, are you using Asiair? I ask it because the problem probably came from the fits file header, and it happen if the meridian flip is activated after you start the Auto mode. 

CS - Álmos
Helpful
andrea tasselli avatar
andrea tasselli avatar
If you flipped crossing the meridian then this might be a reason. Split the integration in batches to find which of the set is giving you grief. Say integrate the first 40, then 80 then 120 and so on. Change the number as you see fit.
Patrick Graham avatar
Patrick Graham avatar
I'm using the integration tool in my PI workflow, not the WBPP.  I haven't used any calibration frames except for the autodark feature in the SkyX pro.  I think you both are correct that the flip may be the culprit.  I hadn't thought of that as PI boasts that their integration tool can factor that into the sequence.  I'll try your suggestion, Andrea.   Also, I usually use the RGGB when debayering but may have inadvertently omitted that selection and it defaulted to Auto.  Thanks for the advice and I'll let you know if the issue is resolved.

CS

Patrick G
Patrick Graham avatar
Patrick Graham avatar
My apologies,  the problem occurs in the star alignment process, not integration.  I've tried several different combination of values without any success.
Well Written
dkamen avatar
dkamen avatar
Hi Patrick,

It must be telling you how many stars it is finding in every image. Look at the process log. Based on your image, you probably want to crank up "Peak Response", perhaps increase detection scale by 1 and maybe decrease Log(Sensitivity) a little bit. 

Cheers,
Dimitris
Helpful Respectful Concise
Almos Balasi avatar
Almos Balasi avatar
In some situation the star align can fail also due the camera bad or hot pixels. You can eliminate this problem in PI with Cosmetic Correction.
andrea tasselli avatar
andrea tasselli avatar
Patrick Graham:
I'm using the integration tool in my PI workflow, not the WBPP.  I haven't used any calibration frames except for the autodark feature in the SkyX pro.  I think you both are correct that the flip may be the culprit.  I hadn't thought of that as PI boasts that their integration tool can factor that into the sequence.  I'll try your suggestion, Andrea.   Also, I usually use the RGGB when debayering but may have inadvertently omitted that selection and it defaulted to Auto.  Thanks for the advice and I'll let you know if the issue is resolved.

CS

Patrick G

I had the issue when trying to align LRGBHa frames to each other and when one (or more) was flipped then PI failed to align. Another extreme case is when less than half of the target frame is contained in the reference frame and the field is poor in stars.
Roger Nichol avatar
Roger Nichol avatar
This is a StarAlignment issue, not ImageIntegration.  You can change your Star Detection settings but more likely your Star Matching where you could increase your RANSAC settings.  I attach my settings which have never let me down with my 550mm scope and ASI2600 camera.

Concise
dkamen avatar
dkamen avatar
To me the stars in that image look slightly out of focus or perhaps just blurry due to problematic transparency. So increasing peak response to say 0.95 should find more stars.
Patrick Graham avatar
Patrick Graham avatar
Thanks, everyone for all the help and suggestions.  I'll try each and every one unitl I can get something to work.  Could be just not enough stars in the field so maybe longer exposures are called for.  Focus should have been ok, but always good to check each time after a slew.  Roger,  I'll try your settings and see what happens.  Thanks for sharing. 


clear skies to all,

Pat
Patrick Graham avatar
Patrick Graham avatar
Well,  after applying all your suggestion and techniques, I finally got the star alignment to run successfully.     Everyone, please accept my thanks for all your help.  I'm still in a steep learning curve here and I appreciate you all sharing your experience and knowledge.

Clear Skies to all,

Patrick G
Well Written Respectful Supportive