Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

HELP! My Planetary Imaging Sucks!

  • Please log in to reply
35 replies to this topic

#1 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 11:37 AM

Greetings, new guy here. I took a stab at Jupiter's opposition last night and thought I did a decent job. Then I saw someone else's (with half the aperture!) and realized I am falling a bit short...by a lot frown.gif

 

Please help me figure out what I'm doing wrong and what I can do to make things better!

rstx2_jupiter_Tv1-10s_1600iso_768x792_20200714-22h05m21s_pipp_g4_ap28.jpg

 

very, very light polluted skies

12" dob (no tracking) and 2x barlow

Canon 1000d w/BackyardEOS captured using live view

1/10th @ 1600iso

Aligned with PIPP, ~2500 frames stacked with autostakkert, and processed with Registax.

 

I'd like to understand the biggest points of failure here. I would think such a massive aperture (12") would produce better images...is it my horrible viewing conditions or dobs just suck at imaging in general? or both?

 

Also, it's challenging to find the correct amount of focus because I have to keep up with the movement while trying to dial it in. This might be an issue as well...


Edited by 06AwzIyI, 15 July 2020 - 11:50 AM.

  • Tulloch likes this

#2 MalVeauX

MalVeauX

    Fly Me to the Moon

  • *****
  • Posts: 7,457
  • Joined: 25 Feb 2016
  • Loc: Florida

Posted 15 July 2020 - 12:00 PM

 

very, very light polluted skies

12" dob (no tracking) and 2x barlow

Canon 1000d w/BackyardEOS captured using live view

1/10th @ 1600iso

Aligned with PIPP, ~2500 frames stacked with autostakkert, and processed with Registax.

 

I'd like to understand the biggest points of failure here. I would think such a massive aperture (12") would produce better images...is it my horrible viewing conditions or dobs just suck at imaging in general? or both?

 

Also, it's challenging to find the correct amount of focus because I have to keep up with the movement while trying to dial it in. This might be an issue as well...

Heya,

 

Just to get the ball rolling:

 

1/10th of a second is too long of exposure, that's 100ms. This will capture seeing turbulence which blurs the image. I would suggest you target 1/100th of a second, or 10ms, as this is fast enough to freeze the seeing so that your lucky imaging has a chance of getting a sharp frame, and not just allowing multiple blurry ones to exist in a single long 100ms capture like what happened with your 1/10th second exposure time. So this is a starting point to help eliminate some blur.

 

Autostakkert!3 can handle the whole process of alignming and stacking just to make it easier on yourself.

 

A big part of getting good data is going to be having good seeing conditions, allowing critical focus, getting thousands of frames, having your mirror thermally acclimated to ambient temperature, and being well collimated. From there, sampling matters. Your Canon 1000D has 5.7um pixels, big ones, which means sampling would need to be a lot closer to F25~F30 (which means a 4x or 5x barlow, not a 2x), but you won't be able to support that without a tracking mount. So doing it from a non-tracking fixed mount, you're probably better off with the 2x or 3x barlow. Your total time on the run is going to be the amount of time it takes for the planet to transit your sensor's FOV on your instrument. You could do this over and over and readjust your mount to have it in the FOV again until you have a couple of hundred to a few thousand frames if possible. Keep it to about 3 minutes total time maybe or less, to avoid rotation problems for now which will increase blur.

 

I think you would have been fine with what you were doing, if it were not for the very slow exposure time resulting in lots of blur from seeing. You already have made a great start! waytogo.gif

 

Very best,


Edited by MalVeauX, 15 July 2020 - 12:02 PM.

  • happylimpet, KiwiRay and 06AwzIyI like this

#3 KiwiRay

KiwiRay

    Viking 1

  • -----
  • Posts: 862
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 15 July 2020 - 12:03 PM

It's one image, so multiple factors could be at play.  The most likely is that seeing sucked.  Where are you located?  Where I am, Jupiter barely gets above 20°, so is always in the thickest part of our atmosphere - conditions have to be near perfect to get a respectable image, and that just doesn't happen much.  Also your camera isn't ideal - with high-speed planetary cameras, we can capture over 20,000 frames in a couple of minutes, giving us a much greater chance of having a large number of frames recorded at those fleeting moments of better seeing.  If you're not sure about focus, spend some time going back and forth until you're confident you've hit the spot - Jupiter has enough clear features that focus in good seeing shouldn't be too hard.  If there's no point at which you can see finer details (even very briefly), then seeing isn't on your side. And don't forget to check collimation!

 

Don't worry about a few bad images.  What you see here are usually people's very best images - everyone has many days when conditions would result in something like what you've captured. 


  • happylimpet and 06AwzIyI like this

#4 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 12:15 PM

Thanks for the help guys, much appreciated!

 

I would suggest you target 1/100th of a second

I actually tried various exposures but it was hard for me to see Jupiter in live view. I'm not sure if I'll be able to pull off 1/100th but I'll try next time.

 

Autostakkert!3 can handle the whole process of alignming and stacking just to make it easier on yourself.

Cool, I didn't know that, thanks for the tip.

 

Your total time on the run is going to be the amount of time it takes for the planet to transit your sensor's FOV on your instrument. You could do this over and over and readjust your mount to have it in the FOV again until you have a couple of hundred to a few thousand frames if possible.

This is actually what I was doing. BackyardEOS makes it somewhat easier because you can move the FOV rectangle around. I was doing this with the 5x zoom feature, which was recommended in the BackyardEOS tutorial video...something about a 1:1 pixel ratio. I was able to stack about 2500 frames this way.

 

I think you would have been fine with what you were doing, if it were not for the very slow exposure time resulting in lots of blur from seeing. You already have made a great start!

 

Thanks for the encouragement and tips!



#5 sg6

sg6

    Cosmos

  • *****
  • Posts: 8,311
  • Joined: 14 Feb 2010
  • Loc: Norfolk, UK.

Posted 15 July 2020 - 12:30 PM

1/10 sec for 2500 frames is a minimum of 250 seconds and Jupiter will have rotated. This is also made worse as I presume that you had to select say only 20% of the final number of frames. Most will have been too poor to be included in the stack.

 

For Jupiter the usual guide is 30fps so 1/30 second or less shot and no longer the 90 seconds end to end.

The identify the best and stack around 200 or so.

In effect only the good frames are included in the stack.

 

In your parameters I would say Jupiter has rotated too much and nothing will place an early image in line with a later one, so the result is a bit of a blur.


  • 06AwzIyI likes this

#6 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 12:34 PM

The most likely is that seeing sucked.  Where are you located?

I'm in Huntington Beach, California...bortle 8/9 :(

Also your camera isn't ideal - with high-speed planetary cameras, we can capture over 20,000 frames in a couple of minutes

Using BackyardEOS, I'm sampling about 17-18 fps. From what I understand, it uses simulated exposure to achieve this. If I wasn't manually tracking, I'd be able to capture way more frames, but I only snagged ~2500 before clouds started rolling in.

 


Don't worry about a few bad images.  What you see here are usually people's very best images - everyone has many days when conditions would result in something like what you've captured.

 

Thanks, I really appreciate it.



#7 KiwiRay

KiwiRay

    Viking 1

  • -----
  • Posts: 862
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 15 July 2020 - 12:39 PM


Experienced imagers here have shown that up to three minutes for Jupiter is fine, so 250 s is a little on the long side (but a minor problem compared to other factors here).  30fps would be considered quite slow - I record 100 fps at a minimum, but you'll be limited by your equipment.  You want to capture enough frames that the number you stack is in the thousands.  I have seen some excellent images captured by people manually tracking big Dobs, but it does make it a lot more challenging!  If you find you have a taste for this hobby, you'll probably soon consider some upgrades to your equipment.


Edited by KiwiRay, 15 July 2020 - 12:42 PM.


#8 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 12:41 PM

1/10 sec for 2500 frames is a minimum of 250 seconds and Jupiter will have rotated. This is also made worse as I presume that you had to select say only 20% of the final number of frames. Most will have been too poor to be included in the stack.

 

For Jupiter the usual guide is 30fps so 1/30 second or less shot and no longer the 90 seconds end to end.

The identify the best and stack around 200 or so.

In effect only the good frames are included in the stack.

 

In your parameters I would say Jupiter has rotated too much and nothing will place an early image in line with a later one, so the result is a bit of a blur.

Interesting! I didn't realize Jupiter rotated fast enough to make that much of a difference. I used the best 66% so this might have something to do with it. Thanks!



#9 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 12:44 PM

Experienced imagers here have shown that up to three minutes for Jupiter is fine, so 250 s is a little on the long side (but a minor problem compared to other factors here).  30fps would be considered quite slow - I record 100 fps at a minimum, but you'll be limited by your equipment.  You want to capture enough frames that the number you stack is in the thousands.  I have seen some excellent images captured by people manually tracking big Dobs, but it does make it a lot more challenging!  If you find you have a taste for this hobby, you'll probably soon consider some upgrades to your equipment.

Alright, yeah I'm around ~18 fps with my Canon 1000d. I guess I need to take her out to really dark skies and see if that is the biggest factor here. I was kind of under the assumption that light pollution didn't matter too much with planetary imaging...


Edited by 06AwzIyI, 15 July 2020 - 12:46 PM.


#10 KiwiRay

KiwiRay

    Viking 1

  • -----
  • Posts: 862
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 15 July 2020 - 12:53 PM

Alright, yeah I'm around ~18 fps with my Canon 1000d. I guess I need to take her out to really dark skies and see if that is the biggest factor here. I was kind of under the assumption that light pollution didn't matter too much with planetary imaging...

It doesn't matter - Jupiter and other planets are so bright that light pollution isn't a factor.  In great, steady seeing, 18 fps can still produce very good images, but it's harder when seeing is more variable, which is much more common. 



#11 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 01:04 PM

It doesn't matter - Jupiter and other planets are so bright that light pollution isn't a factor.  In great, steady seeing, 18 fps can still produce very good images, but it's harder when seeing is more variable, which is much more common. 

"Seeing" meaning how high/low the object is in the sky?



#12 bunyon

bunyon

    Aurora

  • *****
  • Posts: 4,937
  • Joined: 23 Oct 2010
  • Loc: Winston-Salem, NC

Posted 15 July 2020 - 01:05 PM

Just to echo what others have said, "Seeing" has nothing to do with light pollution. It has to do with the steadiness of the air. Local seeing can also have huge effects. You're better off imaging from grass/dirt/sand/etc. You want, as much as possible, to have as little thermal variability as possible between you and Jupiter. You've seen wavy air over a hot highway? The same effect can mar an image made from a hot driveway or parking lot. 

 

Also, that means your optics need to be thermally equilibrated. There are a lot of ways to accomplish that but you need to see to it if you didn't. 

 

And, as others have said, more frames are better. That will be a tough ask with manual tracking but it is possible. 

 

So, shorter exposure, take care to set up such that the scope and ground are thermally equilibrated, focus carefully and, oh, yeah, make sure your optics are perfectly collimated. Then it only comes down to the seeing. If the seeing is bad, none of the other stuff matters much.

 

You mention surprise that images from scopes half your aperture are as good. I would far rather image with a 6 inch under superb seeing than a 25 inch under crappy seeing. I mean, obviously, I'd rather have the 25 inch under excellent seeing but you get the idea. 

 

And, yeah, you're going to have to be willing to face lots of bad images between good ones until you get all these factors under control and/or live somewhere with incredible seeing. But it's a lot of fun!


  • KiwiRay and 06AwzIyI like this

#13 KiwiRay

KiwiRay

    Viking 1

  • -----
  • Posts: 862
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 15 July 2020 - 01:31 PM

Right, sorry for not explaining.  Seeing is related to the steadiness of the air.  As bunyon said, there are a few things you can control (like not imaging with planets over roofs or on warm driveways), but it's mostly beyond your control.  Sometimes the atmosphere above you will be very steady, allowing you to see fine details, while at the worst of times, you'll barely be able to make out the two main dark bands on Jupiter.  Usually it's in between, and often it's quite variable, with moments of steadiness that you want your camera to be fast enough to capture.  And yes, with more atmosphere for light to pass through when objects are close to the horizon, conditions are usually worse the lower a planet is in the sky.

 

As for thermal equilibrium, I put my scope outside 1-2 hours before imaging to allow it to cool to the ambient temperature.  Otherwise you'll see distortion from the thermal currents within the tube.  Some people with larger scopes use fans to speed up or maintain equilibrium.  I'm no expert on this aspect of things, but putting the scope outside for a while has always been enough for me.


  • 06AwzIyI likes this

#14 RedLionNJ

RedLionNJ

    Soyuz

  • *****
  • Posts: 3,898
  • Joined: 29 Dec 2009
  • Loc: Red Lion, NJ, USA

Posted 15 July 2020 - 03:14 PM

You also never mentioned (at least that I saw) where you're located.

 

From mainland US or Europe, for example, Jupiter is horribly low in the skies these days. This greatly affects potential seeing. It also disperses (speads out) different wavelengths of light. You're using a DSLR, so it's basically a one-shot-color (OSC) cam. Users of OSC cams need to use an ADC (Atrmospheric Dispersion Corrector) to counter the natural dispersion at low elevations.

 

You should be able to get BYEOS to go up to around 28 or 29 fps in planetary mode. That's what I run it at with my 550D.

 

When stacking, you really need THOUSANDS of frames.for Jupiter.  Anyone who says otherwise is full of it and shouldn't be listened to.  Particularly here in CN, listen to those who practice, not those who only preach.

 

You'll get there. Seeing, focus, image scale and dispersion are going to be four tough cookies to crack, though. It might take you a couple years or so.

 

BTW - as far as imaging (or even viewing) goes - there's nothing special about opposition - Jupiter (& Saturn) is amply big and visible for 9 months of the year.


  • KiwiRay and 06AwzIyI like this

#15 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 03:29 PM

You also never mentioned (at least that I saw) where you're located.

 

From mainland US or Europe, for example, Jupiter is horribly low in the skies these days. This greatly affects potential seeing. It also disperses (speads out) different wavelengths of light. You're using a DSLR, so it's basically a one-shot-color (OSC) cam. Users of OSC cams need to use an ADC (Atrmospheric Dispersion Corrector) to counter the natural dispersion at low elevations.

 

You should be able to get BYEOS to go up to around 28 or 29 fps in planetary mode. That's what I run it at with my 550D.

 

When stacking, you really need THOUSANDS of frames.for Jupiter.  Anyone who says otherwise is full of it and shouldn't be listened to.  Particularly here in CN, listen to those who practice, not those who only preach.

 

You'll get there. Seeing, focus, image scale and dispersion are going to be four tough cookies to crack, though. It might take you a couple years or so.

 

BTW - as far as imaging (or even viewing) goes - there's nothing special about opposition - Jupiter (& Saturn) is amply big and visible for 9 months of the year.

I'm in southern California and yes, Jupiter doesn't get very high. I'll look into getting an ADC, stacking with more frames, and lowering the exposure. With regard to BYEOS, the most I can push is around 20 fps tops. Not sure if it's just a limitation of my old camera or what :/

 

Jupiter's rotation might be a bigger deal than I initially thought. While 2500 frames @ 1/10th sec exposure does equate to 250 secs, the kicker is that these are not back-to-back shots. Throughout the session I'm moving the scope over and over again trying to center, which takes even more time. It probably took me about 20min to get those 2500 frames from start to finish!


Edited by 06AwzIyI, 15 July 2020 - 03:37 PM.


#16 dhammy

dhammy

    Viking 1

  • -----
  • Posts: 780
  • Joined: 20 Jul 2015
  • Loc: Puerto Rico

Posted 15 July 2020 - 03:44 PM

That's the way we all get hooked - always striving for better and better images smile.gif  

 

Jupiter's rotation might be a bigger deal than I initially thought. While 2500 frames @ 1/10th sec exposure does equate to 250 secs, the kicker is that these are not back-to-back shots. Throughout the session I'm moving the scope over and over again, which takes even more time. It probably took me about 20min to get those 2500 frames from start to finish!

Yeah getting frames from across a 20 min period is going to give you all whole lot of smearing which is what you found with your image. Try a 3 minute video, with a faster exposure time and then stack 50% of those frames to start off with. See what it looks like, then try one with 25% of those frames (not forgetting to reduce your sharpening as you see more noise). Keep experimenting and over time you'll get a good handle on what percentage is a good starting point depending on how good the seeing is.

 

See how your image has a blue and red fringe on opposite sides? Use the RGB align function in Registax - increase the size of the box over the whole planet and then click estimate. That'll reduce the fringing and help with some more detail. 

 

Last night I sat out imaging Jupiter for close to an hour, then I went through all of the individual videos and found only one of them was acceptable due to the turbulent atmosphere. My videos were 2 minutes long each... so that was 2 minutes out of 60 that were worth it! Sometimes you just get lucky, other times there's nothing you can do.


Edited by dhammy, 15 July 2020 - 03:51 PM.

  • KiwiRay and 06AwzIyI like this

#17 rkinnett

rkinnett

    Ranger 4

  • *****
  • Posts: 328
  • Joined: 08 Aug 2018
  • Loc: La Crescenta, CA

Posted 15 July 2020 - 03:52 PM

+1 regarding KiwiRay's explanation of seeing variability and his suggestion regarding letting your scope acclimate.  I just wanted to reemphasize the point about shooting over rooftops or parking lots.  This is one of the few things you can control, and failing to do so is the classic mistake of us beginner planetary imagers. 

 

I came to this thread because I'm also frustrated and disappointed by my own attempts at Jupiter and Saturn last night.  I'm certain the main problem was shooting over my driveway and roof.  Thought I learned that lesson already, but I couldn't leave the house and thought I would try imaging anyway.  I guess it wasn't all in vain, though; it was fun and good practice, and I learned a few things.

 

Getting good, crisp planet images takes a lot of skill as well as equipment capability.  You're rapidly learning everything you need to know and are discovering the limitations of your equipment.  Frankly, I think that image is quite good given the challenges you're dealing with.  It looks like you've figured out wavelet processing in Registax.  You should also use the RGB color alignment feature in Registax to eliminate those red and blue fringes around the upper and lower limbs, and you'll get better contrast overall as well.  Keep it up!

 

I'm no expert in any of this and am struggling to reach the level of detail others are producing with the same or even less capable equipment than mine, but here's my in-work summary of key concepts and capabilities needed for lucky imaging, with some tips I haven't found explained anywhere else:

  • First and foremost: Good seeing.  Need good atmospheric conditions, good target elevation, and don't shoot over hot pavement or rooftops.
  • Scope acclimation:  let your scope acclimate 1-2 hrs before imaging to minimize internal air currents.  By the way, you can check for air currents by pointing at a bright star and defocusing and looking for waviness in the fringe ring.  Make this part of your pre-imaging collimation routine.
  • Collimation must be dead-on.  Any asymmetry will smear your image.  Check collimation prior to every imaging session.
  • Focus must be dead-on.  Use a Bahtinov mask with a nearby bright star.  Check before imaging and re-check periodically.
  • Use an atmospheric dispersion corrector (ADC) to reduce prismatic smearing if you're imaging less than 30 deg or so above the horizon.
  • Use a dew shield and/or heaters to avoid condensation on your optics.  This is less of a problem for Newtonians compared to SCTs and refractors.
  • Use a Barlow to extend focal length for higher magnification, but only if you can do so without over-compromising gain and exposure time.  Also beware, using a focal length extender will decrease contrast and could potentially impart tilt and other adverse effects if not done right.
  • Exposure and gain:  the name of the game is to minimize exposure time in order to maximize your chances of beating seeing.  10ms seems to be a common rule of thumb, but you may not be able to go that low on fainter targets without severely compromising gain.  Exposure times nearing 50ms or higher won't cut it unless seeing is exceptionally good.  I wouldn't hesitate to go to 3200 ISO on my Canon DSLR or gain 300 on my asi224mc in order to minimize exposure time.  In exceptional seeing you can get better results with longer exposure time and lower gain.  What combination works for you depends on your camera, focal length, and aperture.  How do you know if you need more gain and/or exposure time?
  • Check your histogram.  Your combination of gain and exposure should result in the brightest light pixels spanning 50-75% of your camera's dynamic range.  The main peak (representing black space) should be near but not overlapping the left edge, while the right side of the histogram (your brightest pixels) tapers off somewhere above 50% of the range, with some space between that tail and the right edge of the range.  If your histogram is bunched up on the left, either overlapping the left edge or with the brightest pixels less than halfway across the range, then you need to increase gain and/or exposure time.  I generally start with gain until it's as high as I'm comfortable pushing it.  If you still don't have the right balance, gradually increase exposure time.
  • Capture many frames.  Ideally, you want to capture thousands of images so you can be selective in post-processing.
  • Capture those frames quickly.  If you're imaging Jupiter, you have a very short window of only 2-3 mins to capture your entire image set before Jupiter rotates too much for you to stack without smearing, unless you use advanced processing (de-rotation ala Winjupos).  The longer your focal length, the shorter this window becomes since you're reaching for finer detail.
  • Use a stacking program and be selective.  The idea behind lucky imaging is that you capture many images and stack only the best.  Autostakkert seems to be the tool of choice by many admired planetary imagers for initial selection, alignment, and stacking, followed in workflow by Registax for wavelet processing.  Registax can also perform the earlier steps, but I have generally found Autostakkert to be more consistent and easier to use (after you climb the initial learning curve).  Follow online tutorials to learn both.  In both programs, it's important to understand how registration points work, how to evaluate and select a base reference frame, and how to use selectivity.  Regarding selectivity, you could in theory get away with stacking the best 50% or so of your frames (typical starting point for beginners), but if seeing was less than perfect, you're better off bringing that number down.  It's common to stack only the best 5% of your images.  That's why you need to capture thousands of frames to start with.  In Autostakkert, following the Analysis stage, it's essential that you select a good base reference frame, as that will be the frame that all others are compared against to find best matches.  Use the slider in the image preview screen, starting from the left (they are ordered by quality) and find a frame that visually appears to you to be the least distorted and have the most detail.  Then, re-generate your alignment points using this frame as the base.  Experiment with different keep-rates and don't hesitate to push it down to 5% or so, if you have enough source frames to do so.  The image quality histogram can be a useful indicator of how good the most detailed frames are compared to the rest, and you can use that histogram to guide what percentage of frames to keep.  If the curve falls off sharply on the left, then you might be best off with a low keep rate.  If the curve falls off gradually, then most of your frames are similar quality (good or bad) and you could get away with keeping more frames.  Unless seeing was exceptionally good, I like to see a fairly sharp drop off, indicating I indeed captured some frames that are substantially better than the rest.  The width of that peak on the left side of the quality histogram guides my initial selection for keep-rate.

  • SupernovaDust and 06AwzIyI like this

#18 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 04:42 PM

That's the way we all get hooked - always striving for better and better images smile.gif  

 

Yeah getting frames from across a 20 min period is going to give you all whole lot of smearing which is what you found with your image. Try a 3 minute video, with a faster exposure time and then stack 50% of those frames to start off with. See what it looks like, then try one with 25% of those frames (not forgetting to reduce your sharpening as you see more noise). Keep experimenting and over time you'll get a good handle on what percentage is a good starting point depending on how good the seeing is.

 

See how your image has a blue and red fringe on opposite sides? Use the RGB align function in Registax - increase the size of the box over the whole planet and then click estimate. That'll reduce the fringing and help with some more detail. 

 

Last night I sat out imaging Jupiter for close to an hour, then I went through all of the individual videos and found only one of them was acceptable due to the turbulent atmosphere. My videos were 2 minutes long each... so that was 2 minutes out of 60 that were worth it! Sometimes you just get lucky, other times there's nothing you can do.

Yeah I think this was the main issue now. My scope had been outside in my backyard a few hours prior and wasn't shot over a rooftop or anything (grass backyard over a river channel actually), so I think things had decent equilibrium.

 

I'd love to capture a 3 minute video but I can only do about 20 seconds at a time before having to reposition. So I guess there are a couple solutions: get tracking and/or faster frame rates. At this point tracking is out of the question as this scope is just a toy for star hopping mostly and I'm not going to get a mount solid enough to support a 12" dob hah. So I might bite the bullet and pick up a planetary webcam or something.

 

Hmm I might mess around with less frames but choose sets that were captured closer together.


Edited by 06AwzIyI, 15 July 2020 - 04:45 PM.


#19 dhammy

dhammy

    Viking 1

  • -----
  • Posts: 780
  • Joined: 20 Jul 2015
  • Loc: Puerto Rico

Posted 15 July 2020 - 04:58 PM

What you could do is figure out a way to gently nudge the scope to keep Jupiter in the field of view of the camera. If you capture at a faster exposure then this movement won't affect the frames. You could be continually nudging it throughout the 3 minutes - essentially manually guiding it. If you can do it with Jupiter in the eyepiece then you can do it with the camera too. 

 

Perhaps even do a search and see how others have done it - I'm almost certain it's been done.


Edited by dhammy, 15 July 2020 - 04:59 PM.

  • KiwiRay and 06AwzIyI like this

#20 KiwiRay

KiwiRay

    Viking 1

  • -----
  • Posts: 862
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 15 July 2020 - 05:04 PM

Alternatively, you could combine sharpened stacks from several 2-3 minutes videos in WinJUPOS, which allows you to integrate over periods like 20 minutes by derotating the individual images.  But that's usually something people tackle once they've mastered the basics, and I'd recommend trying David's (dhammy) suggestions first.


  • 06AwzIyI likes this

#21 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 05:57 PM

What you could do is figure out a way to gently nudge the scope to keep Jupiter in the field of view of the camera. If you capture at a faster exposure then this movement won't affect the frames. You could be continually nudging it throughout the 3 minutes - essentially manually guiding it. If you can do it with Jupiter in the eyepiece then you can do it with the camera too. 

 

Perhaps even do a search and see how others have done it - I'm almost certain it's been done.

Alright, I'll have to mess around and see if I can come up with something. The problem is that with BYEOS, I'm capturing at 5x zoom which decreases the FOV so it makes it pretty tough. I'm doing this because 5x provides a 1:1 pixel resolution which, supposedly provides better imaging: https://youtu.be/z3gkw8bx7Aw?t=5891

 

I might have to abandon this though, if I can't gather enough frames before jupiter rotates...looks like I have my homework cut out for me boys.

 

Thanks again for all your help, you guys rock!



#22 06AwzIyI

06AwzIyI

    Lift Off

  • *****
  • topic starter
  • Posts: 21
  • Joined: 14 Jul 2020

Posted 15 July 2020 - 06:02 PM

Just out of curiosity, is this (rotation) an issue with Saturn as well?



#23 Achernar

Achernar

    Voyager 1

  • *****
  • Posts: 11,630
  • Joined: 25 Feb 2006
  • Loc: Mobile, Alabama, USA

Posted 15 July 2020 - 06:03 PM

Too slow exposure time, and frame rate. You need to record at a much faster frame rate to freeze the atmospheric turbulence as best you can. Also, you have dispersion caused by the atmosphere, that is why you have the blue and red fringing on opposite sides of the planet.

 

Taras


  • 06AwzIyI likes this

#24 KiwiRay

KiwiRay

    Viking 1

  • -----
  • Posts: 862
  • Joined: 09 Apr 2015
  • Loc: Seattle, USA

Posted 15 July 2020 - 06:12 PM

Just out of curiosity, is this (rotation) an issue with Saturn as well?

Less so, unless you're trying to capture small storms or the polar hexagon.  Otherwise I think you can go up to six minutes with Saturn before field rotation (due to the alt-az mount not tracking with the Earth's rotation) starts to become a problem for Autostakkert.


  • 06AwzIyI likes this

#25 Tulloch

Tulloch

    Surveyor 1

  • *****
  • Posts: 1,557
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 15 July 2020 - 06:54 PM

Hi there, nice topic header, one that is sure to elict a lot of responses smile.gif

 

Firstly, since you are using BackyardEOS and an Canon DSLR, ignore all the comments about using a faster shutter speed, the DSLR cannot handle it. Canon DSLRs in LiveView mode take frames at 30fps, all the LiveView preview settings do (ISO 1600 1/10 sec etc) is simulate what the view would look like if you were using those settings. In actual fact, you are using 1/30 sec at an ISO to match the EV value (probably around ISO 600 with your settings).

 

When I was using my Canon 700D, I was never able to get faster than 20 fps through the USB2 port to my computer even though others were able to get higher with their DSLRs. I believe this is a feature of the camera, and you are unable to get faster speeds than this. I did notice that when I ran my computer off the battery (ie not plugged in), my fps value dropped to about 16-17 fps, so make sure you are plugged in. Some more tips:

 

-  I always recorded at least 5000 frames with my 700D, in two sessions of 2500 jpgs. The avi files that BYEOS creates are massive, the stream of jpgs significantly smaller. Aim to have the maximum histogram value around 50% (so brightest pixel = 127 or so).

-  Learn how to use PIPP to turn those jpgs into a ser file for stacking in AS!3. Center the planet in the middle of the frames and crop them to only include the target. Save as a ser file

-  Use AS!3 to stack the video, not Registax. Stack the best 25% or so, play around to see which gives you best results.

-  Use Registax to sharpen the image using the wavelets tool - try a few different settings

-  Use Photoshop (or your favourite photo editing software) to make final adjustments.

 

For more tips on how to use these programs and create better images, watch these tutorial videos.

http://planetaryimagingtutorials.com/

 

Some of my better images with my DSLR are in my gallery, but here is a summary of some tests I did comparing the Canon 700d with a dedicated planetary camera, the ASI224MC.

https://www.cloudyni...-sct-test-four/

 

Hope this helps,

 

Andrew


Edited by Tulloch, 15 July 2020 - 07:42 PM.

  • 06AwzIyI likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics