Results 1 to 8 of 8
Like Tree1Likes
  • 1 Post By sada

Thread: Planetary imaging over/under sampling?

  1. #1
    skaven's Avatar
    skaven is offline HYPER GIANT
    Points: 35,918, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 4.0%
    Achievements:
    50 Posts Achievement!5 Threads Achievement!200+ Posts Achievement!First 1000 Experience Points5+ Referrals Achievement!
    Join Date
    Dec 2011
    Location
    San Diego
    Posts
    2,656
    Points
    35,918
    Level
    100
    Thanks
    178
    Thanked 1,907x 960 Posts
    Downloads
    1
    Uploads
    0

    Default Planetary imaging over/under sampling?



    I'm trying to figure out the optimal focal length to use when doing prime focus planetary imaging. I've found numerous calculators and articles online, but I'm not seeing a lot of agreement between them.

    I started in Stellarium by entering in my CCD dimensions (4mmx3.2mm) and pixel size (3.125 micron). Using this method, it looks like using a 5x barlow with my scope (1200mm f/4.7) results in a nicely framed Jupiter on the CCD.

    However, running the numbers (using this site: Useful Formulae) shows that with this configuration (6000mm f/23.5) I will be getting an image scale of 0.11 arcsec/pixel.

    It's obvious why undersampling would be bad. With insufficient image scale, I wouldn't be able to record detail the scope is able to resolve, because fine features would fall on too few pixels.

    But is there a danger in oversampling? The sites I've looked at seem to suggest that sampling of around 0.33 arcsec/pixel is about right. But is that just an upper limit? As long as there is enough light to generate a clear image on the CCD, does it matter if I run up the image scale? Obviously I won't be able to resolve things smaller than the Airy disc, but does it matter when doing imaging stuff to magnify significantly beyond the size of the Airy disc? With my 10" scope, the Airy disc is around 1 arcsec in diameter. So with the 5x barlow, the Airy pattern will cover something like 75 pixels. That sounds like a lot, but is it too much? It seems like when it comes to the ability for Registax to extract more and more detail, the more pixels the better.

    But I'd like to hear from the experts on this forum...

    To view links or images in signatures your post count must be 5 or greater. You currently have 0 posts.

    C10-N/C14+Losmandy G11; Portaball 12.5"; NexStar 6SE; Meade ETX125; Stellarvue F80
    Meade Series 5000 HD-60 9mm, 6.5mm; Series 4000 56mm; University Optics Orthoscopic 9mm, 7mm, 5mm;
    ES 70░ 15mm, 20mm; ES 82░ 30mm, 18mm; ES 100░ 9mm; Baader Hyperion 24mm; Morpheus 12.5mm
    William Optics Binoviewer; GSO 2" 2x ED barlow; 5x APO barlow, Parks 2.5x APO barlow
    Filters: Orion Ultrablock, Thousand Oaks OIII

  2. #2
    Phil Leigh's Avatar
    Phil Leigh is offline HYPER GIANT
    Points: 34,455, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 0%
    Achievements:
    5 Threads Achievement!50 Posts Achievement!200+ Posts Achievement!First 1000 Experience Points400+ Posts Achievement
    Join Date
    Nov 2011
    Location
    Cryers Hill, High Wycombe, UK
    Posts
    5,867
    Points
    34,455
    Level
    100
    Thanks
    451
    Thanked 2,014x 1,649 Posts
    Blog Entries
    8
    Downloads
    2
    Uploads
    0

    Default Re: Planetary imaging over/under sampling?

    If the choice is to cover Jupiter with 200 correctly sampled pixels or 100 correctly sampled pixels then yes, 200 is better...

    The key is "correctly sampled" pixels...

    Planetary is not a pixels game, it is a resolution game. Ultimately the resolution of your optice determine what is possible, not the number of pixels that the camera has. This is why the best images come from lowish pixel count cameras... They tend to have larger pixels and pixel size is important for planetary.

    Regardless of camera, You want something like 0.25 arcseconds per pixel for planetary. To get the sampling correct you adjust the focal length of the scope+Barlow. To find the best focal length, multiply the pixel size in microns by 825

    A typical "webcam" has 5.5 micron pixels and so will get best results at a focal length of approx 4500mm, hence the popularity of 2000mm scopes +2/2.5x barlows...

    All of this is limited by seeing of course!

    The best planetary images are taken using 640x480 high frame rate cameras at total focal lengths of 4-10,000mm and focal ratios of f20-f50 with large aperture scopes(the larger the better)

    Registax wavelet sharpenng is not really goverened by the number of pixels, it is governed by the accuracy of the variation in level from one pixel to the next... It is this information that is used and this iwhy getting a good, well filled histogram is important for planetary... To capture the widest dynamic range and subtlest gradation in shading.

    So back to our question ... Is it better to have Jupiter covering 50 or 100 pixels of your camera? the answer is 50 correctly or over sampled pixels are better than 100 under sampled ones.


    Jupiter in the sky is at most 45 arcseconds... 640x480 sensor is more than enough to capture the possible detail even with a 14-inch scope. At 0.25 arcseconds/pixel it would occupy 160 or so pixels.

    There's another practical factor to consider... If Jupiter is so large that it fills the frame of your CCD, you will have the devils own job keeping it on the chip!

    Hope this helps a bit...
    Telescope: SWEquinox ED80 Pro (Schott/Ohara Fluorite)+Baader Steeltrack, C8 XLT Mount: Skywatcher HEQ5 Pro, Rowan belt drive on Pillar mount+EQDIR+ADM saddle/dovetails Camera: JTW1100D mono , CentralDS 600, ASI120MM. ASI120MC, IDAS & Baader NB 2-inch filters, TS Filter Drawer System Accessories: TV Nagler T4 12 mm, Baader 8-24 MkIII, Ortho 5mm, Aspheric 31mm, Meade 12mm reticle, TV Powermate 2x & 2.5x, JMI+BAST Motorfocus+FCUSB, TV 0.8x FR/FF, Celestron 0.63FR, Orion RACI & Mag Mini, Canon IS 15x50 Bins
    Software: EQMOD/Stellarium/Autostakkert!/Registax/BackyardEOS/PixInsight,PHD2,AlignMaster, AstroTortilla, Straton

  3. #3
    skaven's Avatar
    skaven is offline HYPER GIANT
    Points: 35,918, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 4.0%
    Achievements:
    50 Posts Achievement!5 Threads Achievement!200+ Posts Achievement!First 1000 Experience Points5+ Referrals Achievement!
    Join Date
    Dec 2011
    Location
    San Diego
    Posts
    2,656
    Points
    35,918
    Level
    100
    Thanks
    178
    Thanked 1,907x 960 Posts
    Downloads
    1
    Uploads
    0

    Default Re: Planetary imaging over/under sampling?

    Thanks for the reply. Gives me a lot to think about. My camera has a 1280x1024 array of 3.125 micron pixels (at my best estimation -- it's just a webcam, so the specs are hard to find). According to your formula, I need to be shooting for a 3.125 * 825 = 2578mm. That's roughly 2x in my 1200mm C10, and 3x in my 800mm SN102. That suggests that my 5x barlow may be quite excessive.

    What I'm still not clear about though is what "correct sampling" even means. Does it mean taking into account the Airy disc size for your scope and choosing an image scale such that you're not "wasting" pixels by running the image scale up? Or is it something else?

    If an Airy disc "fits" on, say, a 4x4 grid of pixels, that sounds like reasonable image scale. In a 10" scope, the Airy disc is 1 arcsec in diameter, and assuming that 4x4 rule, then Jupiter should be placed on a grid 160x160 pixels, with image scale adjusted per the size of the pixels, which matches up with what you suggested (2578mm).

    Is this the optical principle that backs up this "rule of thumb"?

    Assuming it is, what, then, is the downside of oversampling? If I use an image scale of 0.11arcsec/pixel, then a 1 arcsec Airy disc will fall on a disc of ~75 pixels. That's ~4.5x more image data than is strictly required to get a "properly" sampled image. But is that "bad"? Assuming that I can get the same frame rate, and the histogram looks good, etc. -- will having "too much data" otherwise degrade the image? This is where I'm still confused...

    To view links or images in signatures your post count must be 5 or greater. You currently have 0 posts.

    C10-N/C14+Losmandy G11; Portaball 12.5"; NexStar 6SE; Meade ETX125; Stellarvue F80
    Meade Series 5000 HD-60 9mm, 6.5mm; Series 4000 56mm; University Optics Orthoscopic 9mm, 7mm, 5mm;
    ES 70░ 15mm, 20mm; ES 82░ 30mm, 18mm; ES 100░ 9mm; Baader Hyperion 24mm; Morpheus 12.5mm
    William Optics Binoviewer; GSO 2" 2x ED barlow; 5x APO barlow, Parks 2.5x APO barlow
    Filters: Orion Ultrablock, Thousand Oaks OIII

  4. #4
    skaven's Avatar
    skaven is offline HYPER GIANT
    Points: 35,918, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 4.0%
    Achievements:
    50 Posts Achievement!5 Threads Achievement!200+ Posts Achievement!First 1000 Experience Points5+ Referrals Achievement!
    Join Date
    Dec 2011
    Location
    San Diego
    Posts
    2,656
    Points
    35,918
    Level
    100
    Thanks
    178
    Thanked 1,907x 960 Posts
    Downloads
    1
    Uploads
    0

    Default Re: Planetary imaging over/under sampling?

    As for the "impossible to keep centered" with a full 5x barlow, here's how Stellarium simulates it:

    stellarium-010.png

    The transit time over the CCD is well over 5 minutes -- I feel like it won't be too hard to keep the planet on the chip as long as I do a reasonably good polar alignment.

    Now I just need these darn clouds to go away so I can experiment!!

    To view links or images in signatures your post count must be 5 or greater. You currently have 0 posts.

    C10-N/C14+Losmandy G11; Portaball 12.5"; NexStar 6SE; Meade ETX125; Stellarvue F80
    Meade Series 5000 HD-60 9mm, 6.5mm; Series 4000 56mm; University Optics Orthoscopic 9mm, 7mm, 5mm;
    ES 70░ 15mm, 20mm; ES 82░ 30mm, 18mm; ES 100░ 9mm; Baader Hyperion 24mm; Morpheus 12.5mm
    William Optics Binoviewer; GSO 2" 2x ED barlow; 5x APO barlow, Parks 2.5x APO barlow
    Filters: Orion Ultrablock, Thousand Oaks OIII

  5. #5
    Phil Leigh's Avatar
    Phil Leigh is offline HYPER GIANT
    Points: 34,455, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 0%
    Achievements:
    5 Threads Achievement!50 Posts Achievement!200+ Posts Achievement!First 1000 Experience Points400+ Posts Achievement
    Join Date
    Nov 2011
    Location
    Cryers Hill, High Wycombe, UK
    Posts
    5,867
    Points
    34,455
    Level
    100
    Thanks
    451
    Thanked 2,014x 1,649 Posts
    Blog Entries
    8
    Downloads
    2
    Uploads
    0

    Default Re: Planetary imaging over/under sampling?

    I think you have the optical side of it sorted out. As far as sampling is concerned, as with any digital sampling such as audio, it is always better to err on the side of oversampling. Undersampling loses details. Oversampling consumes additional bandwidth and places higher demands on the entire system. The problem with oversampling in imaging is that it typically means you are operating at a longer focal length = higher focal ratio = less light on the sensor and less contrast in the image. Less light/contrast = less "detail" can be sharpened back in later.

    For your scope on a good night a 2x, 2.5x or even 3x Barlow would be perfect. A 5x will only be useful on nights of exceptional seeing and even then you will still be oversampling.


    As for Stellarium... it will show you accurately how large the image will be - it will give you now idea of how bright/clear it will be. Until you actual try this for real you won't see what I mean.

    In terms of keeping the planet on the chip... Stellarium assumes a perfect alignment and tracking in its simulation. You won't have either. An error of 1 arcsecond in PA will cause the planet to drift at an alarming rate at these magnifications and the higher the mag the faster the drift.

    Try doing a PA drift alignment at 4-5000mm and see what I mean...
    Telescope: SWEquinox ED80 Pro (Schott/Ohara Fluorite)+Baader Steeltrack, C8 XLT Mount: Skywatcher HEQ5 Pro, Rowan belt drive on Pillar mount+EQDIR+ADM saddle/dovetails Camera: JTW1100D mono , CentralDS 600, ASI120MM. ASI120MC, IDAS & Baader NB 2-inch filters, TS Filter Drawer System Accessories: TV Nagler T4 12 mm, Baader 8-24 MkIII, Ortho 5mm, Aspheric 31mm, Meade 12mm reticle, TV Powermate 2x & 2.5x, JMI+BAST Motorfocus+FCUSB, TV 0.8x FR/FF, Celestron 0.63FR, Orion RACI & Mag Mini, Canon IS 15x50 Bins
    Software: EQMOD/Stellarium/Autostakkert!/Registax/BackyardEOS/PixInsight,PHD2,AlignMaster, AstroTortilla, Straton

  6. #6
    sada's Avatar
    sada is offline Banned
    Points: 16,750, Level: 89
    Level completed: 43%, Points required for next Level: 200
    Overall activity: 0%
    Achievements:
    5 Threads Achievement!50 Posts Achievement!200+ Posts Achievement!Ghost Achievement! Averaging 5+ posts a day!400+ Posts Achievement
    Join Date
    Mar 2012
    Location
    Glasgow/Scotland
    Posts
    4,437
    Points
    16,750
    Level
    89
    Thanks
    206
    Thanked 1,002x 963 Posts
    Downloads
    0
    Uploads
    0

    Default Re: Planetary imaging over/under sampling?

    ANOTHER down side to a x5 barlow is any small move meant,a small breeze will be magif 5 times!!!!so you will need a very still night!!!
    Phil Leigh likes this.

  7. #7
    skaven's Avatar
    skaven is offline HYPER GIANT
    Points: 35,918, Level: 100
    Level completed: 0%, Points required for next Level: 0
    Overall activity: 4.0%
    Achievements:
    50 Posts Achievement!5 Threads Achievement!200+ Posts Achievement!First 1000 Experience Points5+ Referrals Achievement!
    Join Date
    Dec 2011
    Location
    San Diego
    Posts
    2,656
    Points
    35,918
    Level
    100
    Thanks
    178
    Thanked 1,907x 960 Posts
    Downloads
    1
    Uploads
    0

    Default

    Thanks guys. Another thing I just realized is that if i set up for "proper" sampling, it means that I must use the camera at its full, native resolution. In my case, that means a piddling 10fps. But if i scale the image up more, I can use the lower resolution options and get better frame rates.

    In the end, it is just going to take some experimentation to get it right. I really appreciate your help!

  8. #8
    Auburn80's Avatar
    Auburn80 is offline Bright Giants
    Points: 2,789, Level: 34
    Level completed: 26%, Points required for next Level: 111
    Overall activity: 0%
    Achievements:
    2 Posts Achievement20 Posts Achievement!5 Threads Achievement!50 Posts Achievement!200+ Posts Achievement!
    Join Date
    Feb 2013
    Location
    Pinson, AL
    Posts
    208
    Points
    2,789
    Level
    34
    Thanks
    33
    Thanked 49x 45 Posts
    Downloads
    0
    Uploads
    0

    Default Re: Planetary imaging over/under sampling?

    Phil,
    Using your math; the Neximage 5 with 2.5 micron pixels would be ideal at 2062 mm. I can get to that easily with a 6" SCT and bang on with an 8" w/o a 2x. Would that work out pretty well - all other factors being nominal?
    Clark
    Scopes: Televue Oracle 3, Celestron C8 SCT, Celestron 6" Newt, Orion/Europa 10" Newt
    Mounts: Celestron AVX, Vixen SP, CG5
    Cameras: ASI120MC, T3i w/ 50mm f1.8, 18-55mm f3.5-5.6, 200mm f2.8L
    Eyepieces: Misc.

 

 

Similar Threads

  1. Using a CCD for planetary imaging
    By pch in forum Astro Imaging Forum
    Replies: 1
    Last Post: 02-11-2013, 09:10 PM
  2. Planetary Imaging
    By isherwoodc in forum Astro Imaging Forum
    Replies: 1
    Last Post: 12-10-2012, 03:01 AM
  3. lunar and planetary imaging
    By Moore se4 in forum Astronomy Forum
    Replies: 0
    Last Post: 10-02-2012, 10:37 PM
  4. Replies: 0
    Last Post: 05-19-2011, 05:18 AM
  5. decent Dob for planetary imaging uses?
    By Wahn Ware Waterston in forum Amateur Astronomy Forum
    Replies: 0
    Last Post: 01-02-2006, 02:58 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Powered by vBulletin« Version 4.2.0
Powered by vBulletin«
All times are GMT. The time now is 11:05 PM.