39 replies to this topic

[johnyj]

 

Rayleigh limits are all about resolving two equally bright objects (such as stars), which is not what we are doing here. Our imaging and processing techniques allow us to do much better than Rayleigh to detect features..

Rayleigh limit is decided by the diameter of Airy disk, which is the smallest information unit you get on an image for a given aperture. In simple terms, if the Airy disk is 9um, covering 3 pixels, then adding a 2x Barlow lens would make it cover 6 pixels, but all those 6 pixels still represent only one valid data point, and is very blurred. 

 

Take this Airy disk photo for example:

 

It took 20 pixels to reveal a star's Airy disk, but that give you a false feeling of an large object, in fact all these pixels represent only one valid data point

 

In Jupiter's case, you could use similar oversampling method, first using a 7x Barlow. Jupiter emit millions of data points, but all of them become an Airy disk after passing the telescope, they stack over each other, turns into a very blurred image, you have to resize the resulting image to regain sharpness. The overall result is no better than just sampling using 3 pixels per Airy disk
 

No amount of imaging and processing could change physical laws, unless that law itself is wrong. But I don't think that is the case, since all modern optical makers build their lens based on these calculations. If a 127mm telescope is really capable of resolving 0.25" details, why don't the telescope maker mark it proudly on their product page? 

Edited by johnyj, 20 September 2023 - 12:22 AM.

[johnyj]

 

celestron 127mm mak f/12 ish  and an asi585 i control it all through an asiair plus and tablet.  i can however use a laptop to record the videos and get much higher frame rates if needed.  i have not tried that yet with this setup and i know it would help having more frames stacked up.  i also track with my orion sirius mount.

 

also mentioned the small bore challenge threads. i have looked and continue to look at those threads and thats what lead me here actually. just trying to make my stuff perform as well as others i have seen.  im also a bit skeptic on pictures posted. i know how easy it is to take a crap pic and photoshop and pretty one on top of it. anyway can fake a photo. lol  but i trust most people dont fake the pics i see on here.

Your photo is already better than my 8" reflector. I also tried shooting video and using post processing software etc... But I could never get close to those photos in small bore challenge threads. That is why I don't really trust them, seems like a resize of an observatory photo. I have friends working at high mountains observatory, even their photos are not that better, for average home observers, I really doubt it unless they all live in Tibet

[CTerry]

Your photo is already better than my 8" reflector. I also tried shooting video and using post processing software etc... But I could never get close to those photos in small bore challenge threads. That is why I don't really trust them, seems like a resize of an observatory photo. I have friends working at high mountains observatory, even their photos are not that better, for average home observers, I really doubt it unless they all live in Tibet

Yeah i hear ya. kinda like looking at a mcdonalds cheeseburger commercial and then going out and getting your own. lol  i think of it more like a hobby or game. something fun to do and waste my time. i dont do it to compete or for any bragging rights or youtuber money or anyhinng like that. just a simple hobby to do in my free time.  so even if i cant make my images look super duper cool like some of the stuff ive seen, im ok with that. just want to keep learning and keep trying for fun.

[Tulloch]

Your photo is already better than my 8" reflector. I also tried shooting video and using post processing software etc... But I could never get close to those photos in small bore challenge threads. That is why I don't really trust them, seems like a resize of an observatory photo. I have friends working at high mountains observatory, even their photos are not that better, for average home observers, I really doubt it unless they all live in Tibet

So just because you've never been able to achieve good results, you accuse everyone else of making up their data?

 

Have a look at this video, it shows me capturing, stacking and processing in real time to an online audience. I captured Jupiter, Saturn and Neptune from my back yard in Melbourne, Australia using 2nd hand consumer equipment.

https://youtu.be/yUPhM2kdxNI?t=81

 

Andrew

[johnyj]

So just because you've never been able to achieve good results, you accuse everyone else of making up their data?

 

Have a look at this video, it shows me capturing, stacking and processing in real time to an online audience. I captured Jupiter, Saturn and Neptune from my back yard in Melbourne, Australia using 2nd hand consumer equipment.

https://youtu.be/yUPhM2kdxNI?t=81

 

Andrew

 

I watched that video, the real time view is more like what I see in my camera live view (4-9x magnified), a photo in post processing does not mean that is the reality, similar to those long exposure nebula photos, because that is not what I see with my eyes in telescope. And, no matter how many frames you process, the resolution of that best frame is still limited by the aperture size, the stacking only solve the seeing problem

 

However, that best frame or one single exposure will very likely end up the same as my real time view in telescope, that is more meaningful

[Tulloch]

I watched that video, the real time view is more like what I see in my camera live view (4-9x magnified), a photo in post processing does not mean that is the reality, similar to those long exposure nebula photos, because that is not what I see with my eyes in telescope. And, no matter how many frames you process, the resolution of that best frame is still limited by the aperture size, the stacking only solve the seeing problem

 

However, that best frame or one single exposure will very likely end up the same as my real time view in telescope, that is more meaningful

Yes, stacking "solves" the seeing problem. Without the atmosphere, what we would see through the telescope should match what we see in the procesed images. The atmosphere smears everything out, sharpening undoes that smearing. If it were all made up, then different people should get different results depending upon the processing they use, but they dont.

 

How do you describe meaningful, what you see yourself through a turbulent atmosphere, or what the object actually looks like?

Edited by Tulloch, 20 September 2023 - 03:55 PM.

[johnyj]

Yes, stacking "solves" the seeing problem. Without the atmosphere, what we would see through the telescope should match what we see in the procesed images. The atmosphere smears everything out, sharpening undoes that smearing. If it were all made up, then different people should get different results depending upon the processing they use, but they dont.

 

How do you describe meaningful, what you see yourself through a turbulent atmosphere, or what the object actually looks like?

By meaningful I mean close to what you see in telescope. Post processing IMO is like art, you could use sharpening and saturation adjustment so that the photo could appear very different than what you see in the telescope

 

In telescope, you can see that Jupiter is a very bright planet, almost white, unlike many of the processed photo here, like an ancient drawing on wall

 

Here is a comparison of my shots lately using no Barlow, 2x, 3x and 4x effective Barlows.

 

These are just single exposure without stacking, but I select the best one out of 20+ shots.

Jupiters on these images are of the size of 50, 100, 150, and 200 pixels. Since my 8" reflector is capable of resolving 0.7 arc second details, that means Jupiter should have 50/0.7= 71 data points, and on Bayer sensor you sample double that amount to be sharp, so 140 pixels should be needed to fully resolve Jupiter. However in these shots, I really don't see any benefit from Barlow larger than 2x, e.g. 100 pixels. On 3x and 4x barlow, the image just get blurry, without more details. Considering I'm using a F5 telescope with 2x Barlow, a F10 telescope would need no Barlow. Maybe this is caused by average seeing, but out of so many shots, the best one is only like this, I think a much better seeing would be very rare

Edited by johnyj, 20 September 2023 - 10:19 PM.

[Tulloch]

By meaningful I mean close to what you see in telescope. Post processing IMO is like art, you could use sharpening and saturation adjustment so that the photo could appear very different than what you see in the telescope

In that case, this is the wrong forum for you. The EAA forum is probably more appropriate, but even that might be too unrealistic for your taste.

https://www.cloudyni...ost-processing/

[Great Attractor]

For example, a guy in Hong Kong just published incredible detailed photo for Jupiter, how is that possible with that bad air in Hong Kong? They just had a typhoon season and typically they can not even see the sun clearly during afternoon (I visit there often)

If by "bad air" you mean poor transparency, it does not impact planetary imaging that much (beside cutting down the brightness).
 

Here is a comparison of my shots lately using no Barlow, 2x, 3x and 4x effective Barlows.

Single frames as above are almost always very poor quality due to a) seeing, b) sensor & photon noise. Both are addressed by lucky imaging+stacking. In addition, thanks to the fact that the target moves randomly about a bit, our final images alleviate pixel grid effects (aliasing) to some extent (quite similar to what “geometric superresolution imaging” on some photographic cameras does).
 

Maybe this is caused by average seeing, but out of so many shots, the best one is only like this, I think a much better seeing would be very rare

Not that rare, as you can see on this forum. If you take thousands of frames, you'll often get 5-10% of good quality ones. Then the noise and pixel aliasing are further reduced by stacking.

 

Think of the process here as what the brain does when you observe for an extended period of time - every now and then there's a spell of sudden image clarity, and your visual cortex remembers for a moment the non-blurred parts of the image. Of course, here we typically also boost the contrast/saturation a bit, which goes beyond the visual view.

Edited by Great Attractor, 21 September 2023 - 05:05 AM.

[johnyj]

The first point may well be valid.

 

The second isn't. "Resolving power" as expressed above is not relevant. People routinely aim for 0.10 arcsec resolution with a C14. Scaling that down to 127mm indicates a resolution of 0.28 arcsec for the 5-inch Mak - almost four times better than the 1 arc second stated.  Take a look at some of the "small bore" threads for empirical evidence of this - there are some mighty fine festoons visible on some of the Jovian images, as well as the darker edges of some white ovals. They're all well under an arc second across, some around 0.25 arcsec.

I did some drawings and it seems what you said for dark patterns in a bright background is true, since dark patterns does not emit light. Even two airy disk bands overlapping each other, if there is no light source under the overlapped area, that area will have less brightness than those area full of light source, thus perceived as a darker line pattern, given enough low exposure

 

Anyway, that is in post processing world, hardly can be seen in a normal day using telescope

[johnyj]

If by "bad air" you mean poor transparency, it does not impact planetary imaging that much (beside cutting down the brightness).
 

Single frames as above are almost always very poor quality due to a) seeing, b) sensor & photon noise. Both are addressed by lucky imaging+stacking. In addition, thanks to the fact that the target moves randomly about a bit, our final images alleviate pixel grid effects (aliasing) to some extent (quite similar to what “geometric super resolution imaging” on some photographic cameras does).
 

Not that rare, as you can see on this forum. If you take thousands of frames, you'll often get 5-10% of good quality ones. Then the noise and pixel aliasing are further reduced by stacking.

 

Think of the process here as what the brain does when you observe for an extended period of time - every now and then there's a spell of sudden image clarity, and your visual cortex remembers for a moment the non-blurred parts of the image. Of course, here we typically also boost the contrast/saturation a bit, which goes beyond the visual view.

I tried that years ago, did not work. I tried it just now again, stacked 4 thousands of frames from a video, and the resulting image is worse than some of my best single exposure shots downsized (video is already downsized to begin with). I don't know what is the calculation behind that staking software (RegiStax), but the result indicating that simply stacking frames would just reduce noise, but no more details

 

 

Previously, I thought that if the seeing is causing distortion in some part of the image,  then it will not always be present, then staking 9 images together would magnify the constantly present signal, but the practice did not show such benefit

[Great Attractor]

stacked 4 thousands of frames from a video, and the resulting image is worse than some of my best single exposure shots downsized

A stack needs to be sharpened. There is always some fuzziness introduced during stacking due to:
 
- somewhat blurry image fragments also getting included in the stack, along with sharp ones
- interpolation of image fragments needed for destretching


 

I don't know what is the calculation behind that staking software (RegiStax), but the result indicating that simply stacking frames would just reduce noise, but no more details

 

I've implemented a stacker, you can read (and see some pics) about what's involved in the algorithms summary. In short, it's not just "simply stacking"/summing pixels, but:

 

1. selection of best (sharpest) frame fragments (lucky imaging)

2. destretching (removing image deformation due to tilted incoming wavefronts)

3. summation (stacking; increases signal/noise ratio)

 

Previously, I thought that if the seeing is causing distortion in some part of the image,  then it will not always be present, then staking 9 images together would magnify the constantly present signal, but the practice did not show such benefit

The practice of everyone else here begs to differ Try sharpening your stacks (via unsharp mask, wavelets, deconvolution...), try imaging on more nights (could have just been unusable seeing).

Edited by Great Attractor, 22 September 2023 - 04:11 AM.

[Winston6079]

 

If what you said is true, then you are defying the theoretical resolving limit of any optical instruments defined by Rayleigh criterion, why not collect more detailed results and submit them to scientific magazines/forums for serious discussion, if that criterion is totally off

 

Based on my own experience, I can not get better photo than OP with my 130mm reflector, and even a 200mm reflector would not help that much, due to average seeing condition

 

I do not really trust others photo here as a reality, so many commercial advertisement. For example, a guy in Hong Kong just published incredible detailed photo for Jupiter, how is that possible with that bad air in Hong Kong? They just had a typhoon season and typically they can not even see the sun clearly during afternoon (I visit there often)
 

I am indeed interested to see the potential of my telescope. A standard way to test optical instrument would be shooting resolution charts in a room and see how many lines per inch you could still separate them on the photo. But since the closest range I can get a focus is at least 100 meters, I have to shoot outdoor targets, that will be difficult for now, so I just trust the theoretical value

Although Hong Kong is usually regarded as a very small city, it is actually about the size of Tokyo. It is also much bigger than Singapore or the Vatican.

 

Some places (such as Yuen Long where famous for the terrorist attack on a railroad station in 2019) have a climate that is unique to the rest of Hong Kong: while the rest of Hong Kong has a sky of dense clouds, it is always cloudless.

 

Yuen Long is one of the rare plains in Hong Kong.

 

As far as I know, almost all of Hong Kong's active astrophotographers come from Yuen Long and the far north of Hong Kong, where houses are low to the ground, making them ideal for photographing planets and even deep-sky objects.

Edited by Winston6079, 06 October 2023 - 10:05 AM.

[RedLionNJ]

Guys, we're getting to be incredibly off-topic, here.  Let's stick to the original post's intent.

[CTerry]

LOL i forgot about this thread.