0.00
#...
·
 |
|---|
|
Hello, looking for an advice please. I am still getting used to a custom build around the babyQ (zwo 2600mm, OAG-L, 36mm filterwheel, flattener 1.01x). I have noticed that sometimes, there are red 'circle' shapes appearing on my images. These do not appear everytime... To me it looks like some chromatic aberration, but I might be mistaken. I also have a very significant vignetting especially on the low edge of my images (this one is noticeable on all the images. I am assuming it is due to the OAG-L used for the guiding camera, but again no certitude here. I am a bit lost and I don't know where to start looking to find the cause of this noise  Any advice is welcome, thanks! P.S. I am using only Darks, no flats or bias frames at the moment.  |
0.00
#...
·
·
1
like
|
|---|
| I don’t see any red spots in your image maybe point us toward them? However you say its noise? If so maybe it’s hot or stuck pixels. Dither during your session and then when you stack the images the hot pixels completely disappear! I have a camera with a lot of stuck pixels and i am using this technique works like a charm! |
|
0.00
#...
·
|
|---|
I don’t see any red spots in your image maybe point us toward them? However you say its noise? If so maybe it’s hot or stuck pixels. Dither during your session and then when you stack the images the hot pixels completely disappear! I have a camera with a lot of stuck pixels and i am using this technique works like a charm! Thank you for taking the time to read and reply. When I mentioned the 'noise', I was thinking of the possibility that these artifacts were due to some thermal noise (dew heater maybe too close to the camera?). These could be a result of eventual lens flares too... As it is the first time I am imaging a bright nebula (M42). Honestly I have no conclusion about what it is... The red circles I describe are indeed not obvious, even less when I take a screenshot of the viewer on a VNC session, sorry for that  Below a properly exported jpeg where these artifacts are a bit more obvious. Following your idea: I noticed they indeed disappear after stacking (Yeah I was too scared something was wrong with my refractor and I submitted to panic :/) And below the stacked output after a bit of postprocessing.  I had to crop hard on the edges to suppress the remains of the vignette. Someone on another channel mentioned that flats are efficient against vignetting. I still need to give that a go. What I find weird about that, is the vignette being much more visible on the lower edge than the upper. So I was wondering if this has something to do with the use of the OAG (compared to using a dedicated guiding scope). |
9.82
#...
·
·
1
like
|
|---|
|
Admittedly, I'm on a phone and it's hard to tell, but I still don't see what you're talking about. You really need to be taking flats and bias however. A proper flat frame will eliminate all vignette. Dark frames are not required on a sensor that does not produce Amp glow. |
7.90
#...
·
·
1
like
|
|---|
Brian Puhl: I very much don't think so. Plenty of examples out there of sensors with no discernible amp-glow which would require dark frame subtraction (lest your image be swamped by thermal noise). |
|
3.01
#...
·
·
1
like
|
|---|
| I see the circles. I wonder is this is a bit-depth problem. What bit depth is your display capable of? |
|
0.00
#...
·
|
|---|
Sean Mc: I'm glad someone can see the same as I see... I was starting to look for an ophthalmologist appointment  My main monitor handles 30-bits, almost double of the 16-bits specified by the camera's manufacturer. Now that you mentioned it, I have checked my stacked image and I notice that it has been saved with 32bits float, instead of 16 bits (as it is for the subs)... Could that be a cause? |
|
7.90
#...
·
·
2
likes
|
|---|
My main monitor handles 30-bits, almost double of the 16-bits specified by the camera's manufacturer. No, it is not ( I guess most of the people here using PI save a 32-bit float) and in fact you should at least save the resulting stacked image as 32-bit integer to preserve the image dynamics. The reason is that you didn't flatten the original image and the background scattered light has a very low bit depth and you have only about 14-bit to play with, so you have this step-like circles. Obviously, because you didn't flatten the raw images, you are already throwing away quite a bit of dynamic range. |
|
0.00
#...
·
|
|---|
andrea tasselli: Thank you for your time. At least it's not a defect affecting my gear Yes I was too lazy to take flats, but note taken! In your opinion, can I cumulate flattened data with non-flattened data of the same subject, when using the same optics and camera? Or should I just throw away the current unflatten data ? |
1.20
#...
·
·
1
like
|
|---|
| You can try to remove stars and use unflatten data, I'm quite sure it's gonna look ok. The effect is usually noticeable on stars. Just make sure you use stars from flattened data. As for the vignetting, you can correct that with some flat frames. I'm using LED tracing pad, works quite well and it was 10 euro. |
|
0.00
#...
·
|
|---|
Robert Žibreg: I will definitely check that pad, thank you ! As for the processing, I did remove the stars but since the core of the nebula is also very bright, I wonder if this makes a difference. It's true however that I don't notice the artifacts anymore after I stack and postprocess, as in the last image  |
|
7.90
#...
·
·
1
like
|
|---|
In your opinion, can I cumulate flattened data with non-flattened data of the same subject, when using the same optics and camera? Or should I just throw away the current unflatten data ? You can flatten your current unprocessed data using tools like Graxpert (GraXpert – A fast and easy way to remove gradients) and then add properly processed data at a later stage. BTW, you are badly burning the core of M42 so probably shorter subs are going to be beneficial. |
|
0.00
#...
·
|
|---|
andrea tasselli:In your opinion, can I cumulate flattened data with non-flattened data of the same subject, when using the same optics and camera? Or should I just throw away the current unflatten data ? |
|
0.00
#...
·
|
|---|
| thank you for your feedback! |
