0.90
#...
·
3
likes
|
|---|
|
Hi everybody, I'm suffering from strong color blindness with a deficiency in the red channel. I posted an image I was kind of proud of yesterday; some of my friends said there was a strong rosy color cast that I was not able to see myself, mixing it up with blue. I did ask my spouse to confirm, which she did  I then tried to remove the cast in PI or in PS to no avail: how can I remove something I cant see? Statistics say thay close to 10% of the male population are affected with color perception deficiency, so I'm hoping to find advice in our community. So how to I usually proceed? I regularly use the "automated" PCC calibration tool to help me with the colors then do my best not to alter them later in the process. This approach has been quite successful on nebulae, according from my mates feedback. But on galaxies, I seem to miserably fail every time. Narrowing down the issue, I suspect that problems arise when merging the Ha layer in the RGB image, independently of the technique used (Ha_AIP, PixelMath or others). I also have calibrated my screen very carefully (SpyderX) and check calibration often. Hence my questions:
Many thanks in advance for your support. KR Rodolphe |
 #...
·
1
like
|
|---|
| I think you should adjust your display to balance your problem. Photoshop has color blindness compensation feature, you could finish editing there |
#...
·
2
likes
|
|---|
| Anyway I don’t dislike the colors in your photos😉 |
10.46
#...
·
4
likes
|
|---|
|
Hi Rodolphe, I'm red blind, so I have a similar situation here and rely solely on PixInsight's tools for color balancing. I'm imaging mostly galaxies and some of my images received comments about great colors (too bad I can't really appreciate the nice colors)  Of course I can't really say whether the color balancing of my images is actually fine. Here is what I use: 1) Color calibration using one of the three tools on the linear image (before stretch): a) BackgoundNeutralisation/ColorCalibration - if the galaxy image can be used as a good while reference. See 'Range Selection Mode' here https://pixinsight.com/doc/tools/ColorCalibration/ColorCalibration.html#description_003 b) PhotometricColorCalibration - for star clusters and if there is no good white reference. c) Recently I've tried AutoColor script from http://www.skypixels.at/pixinsight_scripts.html I've asked a couple of people and it seems that they like the result of the script. 2) SCNR for green on the strethed image. It is done before the final saturation. 3) After SCNR I use HistrogramTransformation to realign the R, G and B peaks to each other in order to make the background neutral. So the final image colors depend only on the image data. I never adjust colors in any other way (because I can't verify the result). Hope this helps. Best regards, Vitali |
|
0.90
#... |
|---|
Alessio Pariani: Hi @Alessio Pariani - I didn't know this feature existed - thanks for pointing me out to it Alessio Pariani: Thanks !!! @sunlover many thanks for your response and detailed steps. I proceed the same way, except for the AutoColor script I didn't try yet, and step 3 (realign RGB peaks) that I'm hesitant to use because I can't check the output. I'll try these steps and report back. KR and clear skies, Rodolphe |
1.51
#...
·
1
like
|
|---|
| In PhotoShop, click on the "set foreground color" square to open color picker and read the RGB values. |
6.08
#...
·
1
like
|
|---|
Rodolphe Goldsztejn: What I have found that helps a lot in doing an Ha-rgb image, is when I have my final stacked Ha-R-G-B files I will combined the Ha and red files to make a kind of super red. What this dose for me is it keeps the Ha from turning pink like it dose if you just layer the Ha in as a lum layer on your rgb image. You still retain the Ha info but it's already a red channel. This works using Photoshop, I don't see why it would not work in PI as well. |
|
0.90
#...
·
2
likes
|
|---|
| @sunlover I used your trick with AutoColor - this is brilliant You may not see the difference, but I confirmed with my spouse that the rosy cast was gone... The Ha integration was the root cause, a challenge for us. See here: https://astrob.in/6mon34/C/ and compare with previous version: https://astrob.in/6mon34/B/ Another trick I will attempt: temporarily swap the blue and red layers - the cast might get more visible? Thanks all for your support. |
|
10.46
#...
·
3
likes
|
|---|
|
Hi Rodolphe. you are welcome, I'm glad that I could help a bit. The AutoColor script seems to be a great tool. It was suggested to me by @Elmiko You are right that I can't say how the colors are different. I see that the stretch level is different Anyway I like the revision C more.BTW, I've used all three methods with my image of the Perseus Cluster: Photometric - https://astrob.in/ssee6m/D/ With a white reference - https://astrob.in/ssee6m/G/ AutoColor - https://astrob.in/ssee6m/H/ The AutoColor revision is the best for me and my daughter also liked it the most Best regards, Vitali |
|
0.90
#...
·
1
like
|
|---|
|
Hi Vitali Your image is incredible! Version H also tends to look best to my eyes. All the best, Rodolphe |
|
10.46
#... |
|---|
| Thanks Rodolphe. |
