2.71
#...
·
 |
|---|
|
Hi, I have been using ZWO ASI 585MC for a few DSO imaging sessions with various filters like UV/IR cut, Orion Skyglow, Optolong L-Pro etc. While using filters I found the stacked images from DSS come out heavily Green tinge saturated. Even when I try to manually adjust the Grey point using Photoshop for RGB alignment the star colors still have weird Red+Magenta tint. What should be the issue? I tried both RGB Alignment & Per channel alignment settings in DSS but issues remain for both options. It becomes difficult for me to fix those star colors even in Siril as it does not recognize stars to do Photometric color correction. Need help to fix the issue of White Balance or Green channel setting for the Camera itself instead of fixing it in post-processing. This is how I get as DSS stacked output -  This is after manual Grey point curve adjustment on the above image  Please help... Thanks |
0.00
#...
·
·
1
like
|
|---|
|
When I used calibration frames this would not happen but I was never happy with the images so I stopped using them. Now I always get a green tint when using lights only but I stack in Siril. I stck by executing scripts, OSC without DBF. You will need to load this script so you can process without calibration frames. After the images are stacked you need to click on auto stretch which will stretch the linear image and introduce all the green. Then go to photometric color calibration. It will populate the Right Ascension and Declination, do not enter the image and click find, just click OK in the window and it will correct the image. After this I run Starnet star removal and process normally. I usually us the auto stretch as it does a better job than me. If it is over stretched you can always move the sliders to dial it back some. I then use Gimp if necessary and then Topaz for denoise and sharpening. Then back to Siril for star recomposition. I hope this helps! I just tried it on the light subs for the Wizard Nebula and it worked perfectly. Aaron |
|
0.00
#...
·
|
|---|
Before Photometric Color Calibration After  |
4.02
#...
·
|
|---|
| (deleted) |
|
0.00
#...
·
·
2
likes
|
|---|
I also just stacked the subs using ASI Deep Stack and they too were green. Then opened the image in Siril and executed the photometric color calibration and it worked after I cropped the image to remove the stacking artifacts which were much more visible the way ASI Deep Stack aligned the images. Maybe you should try cropping the image before color calibration. ages.
|
|
4.02
#...
·
·
1
like
|
|---|
|
In siril, it's super easy to do color calibration. What I do is I select a small box somewhere on the image where there is no DSOs or stars and I color calibrate the whole image using that one spot, then I do a background extraction, and I get perfectly flat image; it works every time. @Aaron ages? ages to you too sir!  |
|
2.71
#...
·
·
1
like
|
|---|
| @Aaron I done as you mentioned. Got this patchy background. Star colors are also more like faint Magenta.  |
|
4.02
#...
·
·
1
like
|
|---|
|
Well, I think you need flat calibration frames. And about the magenta-ish star colors, I think that has to do with both, not having enough integration time, and not doing enough processing; it happens with me and I get stars that brownish or with no color, but I've never gone past one hour integration. I suggest you decrease the brightness of the image to see more clearly the colors of your stars |
6.89
#...
·
·
1
like
|
|---|
|
Sensors record green at twice the intensity of other colours, when you stretch this small difference ends up in an absolute domination of green. This is primarily corrected with background extraction/neutralization, color calibration on its own will also give a more neutral result but it will be unnatural. Calibration subs should have no effect, unless DSS also does background neutralization alongside calibration (I think it does but calls it differently). |
|
4.02
#...
·
·
1
like
|
|---|
Calibration subs should have no effect, unless DSS also does background neutralization alongside calibration (I think it does but calls it differently). I was recommending flats to remove the donuts from his image (assuming they're removable), not fix any colors. Aren't those donuts from sensor dust? |
0.00
#...
·
·
1
like
|
|---|
|
Use background extraction first!!! You don't need any calibration frame on IMX585 (it's a new type of sensor). |
2.11
#...
·
·
1
like
|
|---|
Use background extraction first!!! You "don't" need DARKS with the new sensors (although a still using them with mine QHY268 because of the hot pixels). You defenly need flats flrames, no mether the sensor you use, especially if your sensor have dust spots, and not shooting in a very dark site. |
|
0.00
#...
·
·
1
like
|
|---|
You defenly need flats flrames, no mether the sensor you use, especially if your sensor have dust spots, and not shooting in a very dark site It is much better to clean the sensor/lens than to use flats just for dust spots. It is a very long and tedious way to use the nights. And since dust can move around, doing 3minutes x 25 times flats each time can be really time consuming! I would consider BIAS instead since the IMX585 has a high level of shot noise. But to return to the issue at hand, the green is normal to be evident before erasing the background. |
|
7.90
#...
·
·
1
like
|
|---|
It is much better to clean the sensor/lens than to use flats just for dust spots. It is a very long and tedious way to use the nights. And since dust can move around, doing 3minutes x 25 times flats each time can be really time consuming! No, it's not, even if it this "cleaning" were practicable which isn't. Flats are taken either as dusk/down flats or using a flat panel and the time required in either cases is a tiny fraction of the actual exposures carried out through the session. Most importantly, flats needs taking to correct for PRNU as well as vignetting. Frankly, I'm not reallly even sure we are talking about the same thing here. Bias signal shall not be confused and does not address "shot noise", which is intrinsic to actual process of capturing quantized data, such as photons. Bias correction only addresses the electronic bias error and relevant noise intinsic in powering the detector's electonics. |
|
0.00
#...
·
·
1
like
|
|---|
| Have you tried Graxpert? If not download it, it’s free and works wonders for the background. It might get rid of the dust spots. With plenty of integration time and dithering, I don’t think you need to use calibration frames for this sensor. I haven’t needed them since I refined my process. You will eventually land on what works best for you. I started by processing like the folks on Youtube but I would lose the fine detail in my images so I created my own process and I get great results every time and it usually takes me less than 15 minutes. I use Siril, Graxpert, Gimp and Topaz. |
|
0.00
#...
·
|
|---|
No, it's not, even if it this "cleaning" were practicable which isn't. Cleaning is always a better idea (for a Refractor), and with a tiny sensor like a IMX585 vignetting is CLEARLY not a problem. If you have a Newt IMX585 is not a good Idea. |
0.90
#...
·
|
|---|
andrea tasselli: Agreed wholeheartedly. Even if I spend plenty of time carefully cleaning my sensor, scope lens, etc, etc, there will still be dust on -something- by night's end, especially living in an urban environment. I take my flats immediately after shooting to compensate for anything that has built up on the lens throughout the imaging session, and it makes a substantial difference in image quality during stacking and stretching. You "don't" need DARKS with the new sensors (although a still using them with mine QHY268 because of the hot pixels). Also agreed, here. I find a noticeable difference using darks when shooting in Bortle 8+ conditions. If the sensor doesn't benefit majorly from darks in ideal light conditions, that'll be great in the future (moving to Bortle 4 skies in August). |
|
0.00
#...
·
·
1
like
|
|---|
|
Maybe I'm very fortunate, I have cleaned sensor 2 times, and NEVER cleaned refractor (maybe with a Newt there is lot different experience), and I leave my telescope out the whole night (minimum 9 hours). Dust on sensor with a closed refractor are nearly zero. (with a IMX585 I don't have to switch sensors). But I don't live in a urban environment, in a bortle 4 sky near my house, and a 3 not so far away.  |
|
0.90
#...
·
|
|---|
Maybe I'm very fortunate, I have cleaned sensor 2 times, and NEVER cleaned refractor (maybe with a Newt there is lot different experience), and I leave my telescope out the whole night (minimum 9 hours). To be fair, the dust ends up on the scope lens, and not the sensor, since I'm shooting with refractors and SCTs. Might get a tiny bit of dust when I transfer from the Evostar 72ED to the C6, but I check for it before putting it on the scope. I almost always have some dust/dirt on the scope lens by night's end, depending on wind conditions. Definitely an advantage for you living away from the city! |
|
0.00
#...
·
|
|---|
| SCT with IMX585??????? |
|
0.90
#...
·
|
|---|
| Only for planetary/lunar, but yeah. Haven't noticed any issues with vignette or anything else. |
|
6.89
#...
·
|
|---|
Calibration subs should have no effect, unless DSS also does background neutralization alongside calibration (I think it does but calls it differently). Yes. I was not referring to your recommendation (with which I agree). |
|
6.89
#...
·
·
1
like
|
|---|
Use background extraction first!!! You don't need anything, doesn't mean it won't improve the result.  In this photo about half the subs are with darks while the other half are with bias. There was no noticeable difference. But that was at 0 degrees with a f/6 instrument, and a rich background. Only two days ago I was doing M97 at f/12. Was like "why is it so noisy" (despite the darks), then realisedI had forgotten to switch on the TEC so temperature was 12 degrees. So darks taken at 0 degrees would not correct adequately, and neither would a master bias. I too used to be of the opinion that you can do without calibration subs. There are cases that this is possible, but overall, I was wrong. Of course, they must be done correctly (which can be quite tricky, especially for flats). Otherwise, it is a very real possibility that they will harm the result. But when done correctly, they really do improve it. And it's not like you have to take them every time. In a closed system with controlled temperature, you can go months with the same master calibration subs. Cheers, Dimitris |
|
2.71
#...
·
|
|---|
|
Looks like discussion diverted from main topic. Still not got the solution for removing Green tint from image capture process? |
|
7.90
#...
·
|
|---|
| In Siril is subsky 1 to subtract the background |
