De Bayer pattern for ZWO ASI071MC Pro cmos camera

[deepsouthernskies]

Hi,
I am presented with a dilemma in using the correct De Bayer pattern for the above mentioned color cmos camera. Using Auto produces a yellow image. All the others are also incorrect except for two: RGBG and BGRG. Both of these produce natural color but do not know which one is the correct one. Can someone help?
Thanks, Al

[WillB]

Hi Al

Yellow is not abnormal for an OSC image obtained under less than perfect skies.

Click the link RGB button in the STF tool to deselect it and separate the colour channels and then click the auto stretch button again, now the yellow cast should be gone.

To verify an unknown Bayer pattern the easiest way is to point the camera at a daylight lit primary colour object (red blue or green) so that the object fills the field of view, take an image and process it. With the link RGB channels button deselected in the STF tool only one Bayer pattern will produce the correct colour.

The camera does not need to be on the telescope, cooled or used with a lens for this test just placed close enough to the object so that it fills the FOV.

You will probably find this a faster solution than waiting for a reply from another user with the same camera.

The actual Bayer pattern used by the camera manufacturer's hardware driver and your capture program will be embedded in the image FITS header and may be read using the FITS header tool, this is the source of the Bayer pattern used when you select auto in debayer.

William.

[bulrichl]

I am using an ASI071 as well, with the ASCOM camera driver and SGP as capturing software. The FITS files saved by SGP don't contain the Bayer/mosaic pattern. So I have to set 'Bayer/mosaic pattern' for the Debayer process explicitely to 'RGGB', or the error message

*** Error: Unable to acquire CFA pattern information: Required image metadata not available.

will appear. Note that there is no standardized FITS keyword at all for the Bayer/mosaic pattern! However, some capturing software might use non-standard FITS keywords for this purpose (i.e. BAYERPAT, XBAYROFF, and YBAYROFF).

The correct Bayer/mosaic pattern for the ASI071 is 'RGGB' (assuming both XBAYROFF and YBAYROFF are zero) or 'GBRG' with XBAYROFF = 0. and YBAYROFF = 1.

Anyway your findings are very interesting and might indicate a little bug in PixInsight. So please take a look at a FITS file with the 'FITSHeader' process (from the menu: FILE | FITS Header): determine whether these 3 non-standard keywords are used and which values are assigned. Please present the results here (screen section of the FITSHeader process would be fine).

Bernd

P.S.: Please also note the used camera driver and capturing software.

[Alfredo Beltran]

The correct Bayer/mosaic pattern for the ASI071 is 'RGGB'

Exactly. As Bernd says, the correct pattern is RGGB. I´ve always used it successfully with the 071 Pro.

Regards,

Alfredo

[deepsouthernskies]

Hi, Thanks for the replies.
 I have made a beginner's mistake! This is first light with my OSC cmos camera and the first time I have used PI for image processing. The images were binned 3x3 because the pixels are only 4.78um and I figured I would be way oversampling otherwise. However, I cannot debayer binned images and have to process it as monochrome.
I checked the FITS header and the debayer pattern is RGGB as already described.
This leads me to this question. Would it be appropriate to take binned images to use as a luminance for the color images that are unbinned?
Thanks, Al