Red colors are fainting

[alexander.horn]

Hello, I am trying to use pixinsight, and use the approach described in "Insight Pixinsight" on M42 shot with a Celestron C11 with Hyperstar, Nikon Z6, 20 sec, 50 Lights.
I am wondering why the initial images, as taken from my camera as RAW and imaged in Photohop, has a lot of red colors, see first image, and then processing with Pixinsight, the colors are lost (or not?). I used the debayer as well, as described in the book, I am wondering.
I am sure, I am doing something wrong, but do not see it ... please give me the right idea

 

[pfile]

likely you need to do some kind of color calibration which i'm sure is covered at length in Inside Pixinsight...

rob

 

[alexander.horn]

Dear Rob, thank you. I am wondering, because I had the same problem starting in past with DSS (deep sky stacker). I can imagine that using some color calibration I can somehow get similar colors. But this is not my intention. The information about the amount of colors is given in the RAW images, and somehow they are lost or wrong transformed into other colors when processed in pixinsight.
I expect that the debayer process should there be in charge.

 

[FlapAstro (aka rakla)]

Maybe you use an incorrect depayer pattern? Try to determine the correct pattern for your camera / sensor and force the debayer process to use that instead of ’auto’.

/Ralf

 

[pfile]

yes wrong debayer settings are possible.

as for the color, pixinsight ignores all the color balance information in the CR2 files (when set to Pure Raw) because that info is really only correct for daylight photos. i am not sure what DSS does.

rob

 

[alexander.horn]

Thank you Rob and Ralpf!
Maybe I am thinking wrong, but does the Nikon over-exagerate the red colors? I did foolwing test: I took one RAW image and opend it in PI with the RAW-settings: Demoisaiced RGB. Then I saved this image as TIFF, opend it in Photoshop, andn processed it. See image "Picture_RAW2PIwithDEBAYER2PS".
Then I took the same RAW image and opend it in PI with the RAW-setting "Pure raw". I processed it in PI with Debayer using the settings Bayer/moiaic pattern RGGB, and VNG for demosaicing method. no evaluation of noise. Then I saved it as TIFF and processed same as before with PS. See fig. Picture_RAW2PIwithoutDEBAYER_inPI_DEBAYER2PS. Where are the red colors gone?

 

[pfile]

can you post one of your NEF files? hard to say what's going wrong without looking at the data. you will need to upload it to dropbox or google drive or similar and post the link here.

rob

 

[alexander.horn]

Dear rob,

her is the NEF-File : https://www.dropbox.com/s/8oqkakijh5y8fa0/L_7625_ISO500_20s__NA.NEF?dl=0
Thanks,

Alexander

 

[pfile]

with PI's RAW handler set to "pure raw" i opened the NEF and debayered it with auto pattern. i then had to rotate it. i also mirrored both of them to put the nebula right-way-round. next i opened the NEF file with the raw handler set to "demosaiced RGB".

after that i used ImageSolver to solve the PI-debayered file and then ran PhotometricColorCalibration and the colors remained the same (after recomputing the STF with the channels locked.)

in this screenshot the "demosaiced RGB" file is on the left and the PI-debayered file is on the right. the "demosaiced RGB" file is shown with an unlocked-channel STF and as mentioned above the PI-debayered file has channels locked since it has undergone color calibration.

so in short i don't see any problem here...

rob

 

[alexander.horn]

Dear Rob, thank you. I have to learn a lot about PI. At the meoment I was able to repeat all the steps you descrived, but relly do not understand them. Especially the PhotometricColorCalibration and the ImageSolver are at the moment "magic".
But yes, the images look the same and I can start, knowing that everything is fine
Yours sincerely,

Alexander

 

[alexander.horn]

Hello! I just took the initial image I got from PI (see my first post herin) and applied PhotometricColorCalibration and the ImageSolver ... I used some PS and here it is. Thank you all for you help!