Trouble processing ...
 
Share:

Trouble processing OSC images of M 20  

  RSS

(@thompeters)
Main Sequence Star Customer
Joined: 2 years ago
Posts: 39
August 4, 2018 21:17  

I have about 2.5 hours of imaging on M 20 now from 2 sessions. I'm using a Starwatcher 100 mm APO, with a flattener/reducer, an IR/UV cut filter, and a ZWOASI1600MC, (a single shot color camera, uncooled).

The multi session processing works great and I used it for this image. I also processed each sessions data separately. 

I can not get any other kind of image with this data other than a green image. I've done light pollution removal, background neutralization and star calibrations, and in the end nothing but a green image. I've tried several different ways of doing the star color calibration to no avail.

This is a small .jpg of the integrated stack after many trials. 

Please tell me what I'm doing wrong or not doing!?!?!?! I've never had this problem with such a green image before!!!

I'VE PROCESSED MANY OTHER IMAGES AND HAVE NEVER HAS A PROBLEM WITH GREEN LIKE THIS ONE!!

St avg 9180.0s WC 1 3.0 none x 1.0 LZ3 NS full qua add sc BWMV nor AA RL noMBB session 1 session 2 cbg St Cropped

This topic was modified 11 months ago 2 times by thompeters

ReplyQuote
(@gregwrca)
Neutron Star Customer
Joined: 2 years ago
Posts: 223
August 6, 2018 06:07  

Check your debayer pattern. If I don't force GRBG my colors are all off too

GW


ReplyQuote
(@thompeters)
Main Sequence Star Customer
Joined: 2 years ago
Posts: 39
August 6, 2018 10:59  
Posted by: gdwats@comcast.net

Check your debayer pattern. If I don't force GRBG my colors are all off too

Thank you sir!! Mistakenly I had set the bebayer to RGGB!!! No wonder it was so green!! LOL


ReplyQuote
(@gregwrca)
Neutron Star Customer
Joined: 2 years ago
Posts: 223
August 6, 2018 22:50  

Glad I could help  👍 

GW


ReplyQuote
(@thompeters)
Main Sequence Star Customer
Joined: 2 years ago
Posts: 39
August 7, 2018 01:11  

So I reran my integration this time using the GRBG setting for deBayer. I got a different result, but one I still can not tease the colors I normally see in photos of M 20. I'm super happy with the nice round stars. This image has been through "remove light pollution, Neutralize Background, and Star color Calibration.

I've tried changing the parameters in star color calibration with no success. Also in HSL selective color. Nothing seems to work.

Again try as I may I can not tease out the "correct" colors. Not in APP or Photoshop CC.

SMH!!

St avg 13500.0s WC 1 3.0 none x 1.0 LZ3 NS full qua add sc BWMV nor AA RL MBB20 7 1 18 8 4 18 cbg csc First

 

Thom

This post was modified 11 months ago by thompeters

ReplyQuote
(@mabula-admin)
Quasar Admin
Joined: 2 years ago
Posts: 2081
August 7, 2018 13:00  

Hi Thom @thompeters & Greg @gregwrca,

I agree, after having set the correct Bayer pattern, the result still looks odd.

Perhaps I need to have a look at a single exposure of the asi1600MC. Can you send me 1 or a couple of exposures, so I can have a look? Perhaps send me the integration result that you have as well.

So only a IR/UV cut fitler was used?

Kind regards,

Mabula

 

Main developer of Astro Pixel Processor and owner of Aries Productions


ReplyQuote
(@thompeters)
Main Sequence Star Customer
Joined: 2 years ago
Posts: 39
August 7, 2018 16:41  

Ok Mabula! Remind me on how to send you the files??

yes, only the UV/IR cut filter used. 

Thom


ReplyQuote
(@thompeters)
Main Sequence Star Customer
Joined: 2 years ago
Posts: 39
August 7, 2018 19:15  

OK Mabula - I found the old message so I'm sending the files by wetransfer.com to your support email (support@astropixelprocessor.com).

 

Thanks Thom


ReplyQuote
(@mabula-admin)
Quasar Admin
Joined: 2 years ago
Posts: 2081
August 7, 2018 21:30  

Hi Thom @thompeters,

Thanks for the files, I wll report back my findings as soon as possible 😉

Mabula

Main developer of Astro Pixel Processor and owner of Aries Productions


ReplyQuote
(@thompeters)
Main Sequence Star Customer
Joined: 2 years ago
Posts: 39
August 17, 2018 11:06  

Just wondering if you'd had time to look at the files yet Mabula?


ReplyQuote
(@mabula-admin)
Quasar Admin
Joined: 2 years ago
Posts: 2081
August 19, 2018 22:44  

Hi Thom @thompeters,

Not yet, apologies for the delay.

Can you please send me the data again by wetransfer.com ? (Download has expired)

I will then download it immediately, I promise 😉

Mabula

 

 

Main developer of Astro Pixel Processor and owner of Aries Productions


ReplyQuote
(@thompeters)
Main Sequence Star Customer
Joined: 2 years ago
Posts: 39
August 20, 2018 00:06  

I know you’re super busy, no problem!!


ReplyQuote
(@mabula-admin)
Quasar Admin
Joined: 2 years ago
Posts: 2081
August 20, 2018 15:19  

Dear Thom @thompeters,

I have checked your data. Something odd is happening with the captures of SGP.

The 2 images that you sent me requires different CFA patterns.

M 20 Trifed Nebula_120sec_gain_NA_offset_NA_Deg NA__frame13.fit

FITS header:

FITS HDUs: 1
HDU1 - SIMPLE = T / file does conform to FITS standard
HDU1 - BITPIX = 16 / number of bits per data pixel
HDU1 - NAXIS = 2 / number of data axes
HDU1 - NAXIS1 = 4656 / length of data axis 1
HDU1 - NAXIS2 = 3520 / length of data axis 2
HDU1 - BZERO = 32768 / offset and data range to that of unsigned short
HDU1 - BSCALE = 1 / default scaling factor
HDU1 - CRPIX1 = 2328 / reference spectrum pixel coordinate for axis 1
HDU1 - CRPIX2 = 1760 / reference spectrum pixel coordinate for axis 2
HDU1 - CTYPE1 = 'RA---TAN' / standard system and projection
HDU1 - CTYPE2 = 'DEC--TAN' / standard system and projection
HDU1 - OBJECT = 'M 20 Trifed Nebula' / Object name
HDU1 - DATE-LOC= '2018-08-04T01:01:26' / Local observation date
HDU1 - DATE-OBS= '2018-08-04T05:01:26' / UTC observation date
HDU1 - IMAGETYP= 'LIGHT ' / Type of frame
HDU1 - CREATOR = 'Sequence Generator Pro v3.0.2.91' / Capture software
HDU1 - PROFILE = 'ASI1600 & AVX' / SGPro profile name
HDU1 - INSTRUME= 'ZWO ASI1600MC' / Instrument name
HDU1 - EXPOSURE= 120 / Exposure time in seconds
HDU1 - XBINNING= 1 / Camera X Bin
HDU1 - CCDXBIN = 1 / Camera X Bin
HDU1 - YBINNING= 1 / Camera Y Bin
HDU1 - CCDYBIN = 1 / Camera Y Bin
HDU1 - TELESCOP= 'Celestron Telescope Driver' / Telescope name
HDU1 - RA = 270.618711205562 / Object Right Ascension in degrees
HDU1 - DEC = -22.9920816105201 / Object Declination in degrees
HDU1 - CRVAL1 = 270.618711205562 / RA at image center in degrees
HDU1 - CRVAL2 = -22.9920816105201 / DEC at image center in degrees
HDU1 - OBJCTRA = '18 02 28.491' / Object Right Ascension in hms
HDU1 - OBJCTDEC= '-22 59 31.494' / Object Declination in degrees
HDU1 - AIRMASS = 3.1485563844206 / Average airmass
HDU1 - OBJCTALT= 18.4023251531017 / Altitude of the object
HDU1 - CENTALT = 18.4023251531017 / Altitude of the object
HDU1 - ANGLE = 0 / Image angle
HDU1 - SCALE = 1.02 / Image scale (arcsec / pixel)
HDU1 - PIXSCALE= 1.02 / Image scale (arcsec / pixel)
HDU1 - GAIN = 139 / Camera gain
HDU1 - EGAIN = 4.96000003814697 / Electrons Per ADU
HDU1 - OFFSET = 20 / Camera offset
HDU1 - END

Check below screenshots, I just set a different Bayer pattern and reloaded the image in the image viewer:

RGGB:

M20 frame13 RGGB

GBRG:

M20 frame13 GBRG

BGGR:

M20 frame13 BGGR

GRBG:

M20 frame13 GRBG

When compared to this image from Diego Colonnello, it clearly seems that GBRG is the correct pattern for this image:

Zoom 100 percent

But when we check the other frame

M 20 Trifid Nebula_180sec_Bin 1x1_Gain 139_ASI Camera (1)_24C_0011.fit

Fits header:

FITS HDUs: 1
HDU1 - SIMPLE = T / file does conform to FITS standard
HDU1 - BITPIX = 16 / number of bits per data pixel
HDU1 - NAXIS = 2 / number of data axes
HDU1 - NAXIS1 = 4656 / length of data axis 1
HDU1 - NAXIS2 = 3520 / length of data axis 2
HDU1 - BZERO = 32768 / offset and data range to that of unsigned short
HDU1 - BSCALE = 1 / default scaling factor
HDU1 - CRPIX1 = 2328 / reference spectrum pixel coordinate for axis 1
HDU1 - CRPIX2 = 1760 / reference spectrum pixel coordinate for axis 2
HDU1 - CTYPE1 = 'RA---TAN' / standard system and projection
HDU1 - CTYPE2 = 'DEC--TAN' / standard system and projection
HDU1 - OBJECT = 'M 20 Trifid Nebula' / Object name
HDU1 - DATE-LOC= '2018-07-01T03:05:56' / Local observation date
HDU1 - DATE-OBS= '2018-07-01T07:05:56' / UTC observation date
HDU1 - IMAGETYP= 'LIGHT ' / Type of frame
HDU1 - CREATOR = 'Sequence Generator Pro v2.6.0.25' / Capture software
HDU1 - PROFILE = 'ZWO ASI1600' / SGPro profile name
HDU1 - INSTRUME= 'ASI Camera (1)' / Instrument name
HDU1 - OBSERVER= 'Thom Peters' / Observer name
HDU1 - SITENAME= 'Vicksburg, MI' / Observatory name
HDU1 - SITEELEV= 265 / Elevation of the imaging site in meters
HDU1 - SITELAT = '42d8m10.000s N' / Latitude of the imaging site in degrees
HDU1 - SITELONG= '85d33m8.000s W' / Longitude of the imaging site in degrees
HDU1 - EXPOSURE= 180 / Exposure time in seconds
HDU1 - CCD-TEMP= 24.2 / Camera cooler temperature
HDU1 - XBINNING= 1 / Camera X Bin
HDU1 - CCDXBIN = 1 / Camera X Bin
HDU1 - YBINNING= 1 / Camera Y Bin
HDU1 - CCDYBIN = 1 / Camera Y Bin
HDU1 - XPIXSZ = 3.8 / Pixel Width in microns (with binning)
HDU1 - YPIXSZ = 3.8 / Pixel Height in microns (with binning)
HDU1 - ANGLE = 0 / Image angle
HDU1 - SCALE = 0 / Image scale (arcsec / pixel)
HDU1 - PIXSCALE= 0 / Image scale (arcsec / pixel)
HDU1 - GAIN = 139 / Camera gain
HDU1 - EGAIN = 1.0011096841452 / Electrons Per ADU
HDU1 - END

the GRBG pattern is needed.

RGGB:

M20 0011 RGGB

GBRG:

M20 0011 GBRG

BGGR:

M20 0011 BGGR

GRBG:

M20 0011 GRBG

So this is rather odd to say the least. Since I assume that it's shot with the same camera, the explanation must be found in the capture software SGP Pro. Apparently, GRBG needs to be the correct pattern since I can read that on several sources on the internet if I google.

GBRG then can only be correct if the capture software did a flip of the captured data in the y-axis when storing the data.

Somehow, SGP Pro must provide doing this in the metadata, but I can't find it. I do see that different SGP versions were used:

M 20 Trifid Nebula_180sec_Bin 1x1_Gain 139_ASI Camera (1)_24C_0011.fit was shot with

HDU1 - CREATOR = 'Sequence Generator Pro v2.6.0.25' / Capture software

and

M 20 Trifed Nebula_120sec_gain_NA_offset_NA_Deg NA__frame13.fit

HDU1 - CREATOR = 'Sequence Generator Pro v3.0.2.91' / Capture software

By looking at the other metadata tags, clearly different settings were used in SGP per session.

What I would advise in this particalur case is:

  • keep the data separate per session.  Don't use the multi-session mode.
  • set the correct CFA pattern for the session and calibrate the data, save the calibrated data with the option, align channels. The saved frames are then already debayered properly.
  • After having done this per session, you should have all frames calibrated and propely debayered.
  • Load the calibrated, debayered frames as lights and process as usual.

That should do it 😉

What happened: since the different sessions needed a different CFA pattern, whichever pattern you chose, part of the frames were debayered with a wrong pattern which actually have opposing colors in Red and Blue, which also explains that the end results became quite colorless/green. The different sessions more or less neutralized the Red and Blue color of the other session... but green was maintained.

I hope this explains the issue. It's very odd and SGP Pro or some setting in SGP Pro is clearly to blame here if all data was shot with exactly the same camera but different settings in SGP, or different SGP version.

Mabula

This post was modified 10 months ago 3 times by Mabula Haverkamp - Admin

Main developer of Astro Pixel Processor and owner of Aries Productions


ReplyQuote
(@thompeters)
Main Sequence Star Customer
Joined: 2 years ago
Posts: 39
August 20, 2018 17:18  

Thanks!!!

 

Indeed there was an update to SPG, and I did not have the ASCOM version of the ZWO ASI1600 camera loaded for the first session.

 

What I think I'm going to do is delete all this data and reshoot with the new version of SPG and the ASCOM driver for the ASI1600 camera loaded. And I'll use the GRBG CFA selection for debayer.

 

I did shoot the M 13 Cluster with the new version of SPG, and the ASCOM driver for the camera loaded.  This is the result:

M 13 Great Globular Cluster in Hercules   6.8M

This post was modified 10 months ago 3 times by thompeters

ReplyQuote
(@mabula-admin)
Quasar Admin
Joined: 2 years ago
Posts: 2081
September 6, 2018 23:23  

@thompeters,

Glad we got to the bottom of this 😉 I can really understand the confusion from mixing the data from both sessions which need a different CFA pattern.

Nice M13 !

Mabula

Main developer of Astro Pixel Processor and owner of Aries Productions


ReplyQuote
Share: