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

Region of Interest - new fov strategy?

  • Please log in to reply
8 replies to this topic

#1 cfichter

cfichter

    Vostok 1

  • *****
  • topic starter
  • Posts: 168
  • Joined: 04 Aug 2016

Posted 19 February 2020 - 03:55 PM

I've been imaging almost 10 years now, various experience with APOs and SCTs.

 

I'm really interested to hear folks thoughts about these new full frame CMOS sensors (the IMX455) coming in at an approachable (well still insane for a hobby) price range. Yes I know CCD cameras have had much larger sensor capabilities in the past, but that was a pretty limited audience at $20k camera price ranges...

 

In the past, it felt like I had to chase FoV with different scopes of different focal lengths.  This sensor, using 'Region of Interest' settings in capture software (if I'm understanding correctly) will now allow me to basically set field of view per target (obviously up to the limit of the sensor size) and suddenly I have a wide range of FoV to choose from with the same scope. Albeit, I'll have different file sizes as I use more of the sensor.

 

Is anyone doing this actively, and how practical is this?  With the various capture programs, how easy/reliable can you use Region of Interest and does it over-complicate things -

 

I'm thinking of much smaller file sizes for smaller targets is a huge advantage, but if just too complex, meh, just crop and be done and take the hit on disk storage.

 

Interested in thoughts and experience so far -



#2 Madratter

Madratter

    Voyager 1

  • *****
  • Posts: 11,957
  • Joined: 14 Jan 2013

Posted 19 February 2020 - 04:08 PM

This would require you to make sure you had a telescope that will illuminate the large image circle AND have decent quality throughout that large image circle. Those things don't come cheap.



#3 johnpane

johnpane

    Vostok 1

  • *****
  • Posts: 125
  • Joined: 13 Dec 2014
  • Loc: Wexford, PA, USA

Posted 19 February 2020 - 04:34 PM

File size seems like the least of our worries. I say, capture the whole frame and decide later about whether and how much to crop.


  • MikeMiller, DSOs4Me and OldManSky like this

#4 TOMDEY

TOMDEY

    Fly Me to the Moon

  • *****
  • Posts: 5,939
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 19 February 2020 - 04:54 PM

The concept and practice have been around for a long time, on the professional side of the fence, where a firehose of data gets sifted for pertinence before transmission, else flood the datalink with overflowing useless junk. In the context of astro image-capture... profoundly slow and sparse, in comparison... it's generally considered prudent to collect it all and process/downselect after the fact. Actually, there's a profoundly anthropic element to this interesting subject. Indeed, we are consciously aware or only a miniscule portion of the data streaming in from our senses. Only the pertinent stuff enters awareness, usually for some Executive Decision. That's what ~smart~ image-capture intends to emulate.    Tom



#5 klaussius

klaussius

    Ranger 4

  • *****
  • Posts: 356
  • Joined: 02 Apr 2019
  • Loc: Buenos Aires

Posted 19 February 2020 - 05:11 PM

File size seems like the least of our worries. I say, capture the whole frame and decide later about whether and how much to crop.

Proper ROI settings also reduce download time (during capture I mean) which, I've heard, is a significant source of wasted time for some.

 

With huge FF sensors approaching 40MP or more, it may be significant.


Edited by klaussius, 19 February 2020 - 05:12 PM.


#6 Tulloch

Tulloch

    Viking 1

  • *****
  • Posts: 895
  • Joined: 02 Mar 2019
  • Loc: Melbourne, Australia

Posted 19 February 2020 - 06:03 PM

Using correct ROI and cropped areas is an essential part of planetary imaging to ensure high frame rates and reduced imaging size. Programs like Firecapture can track the target object across the screen so it always stays in the FOV, this might work for DSO imaging also.

 

Andrew


  • MikeMiller likes this

#7 cfichter

cfichter

    Vostok 1

  • *****
  • topic starter
  • Posts: 168
  • Joined: 04 Aug 2016

Posted 20 February 2020 - 08:03 PM

I meant to add a bit more about this in my original post -

 

File size is an issue, I'm setting up a remote observatory rig in Chile - so, ya, I care about file size.

 

What I meant to ask - how robust is 'region of interest' in capture software?  Are you limited to only specific sizes, or can you more or less set your own scale?

 

I've been an SGP user for a long time, going to give N.I.N.A. a try this Spring.  I've heard CCDAutopilot support region of interest well but I'm not familiar with that software.

 

Just interested to hear if others are using this approach and if it is reliable, or just too cumbersome.

 

yes, I'm well aware of image circle requirements on the scope and overall image train - this is really a question about capture software support and direct chip-awareness for scale settings if that makes sense.



#8 Madratter

Madratter

    Voyager 1

  • *****
  • Posts: 11,957
  • Joined: 14 Jan 2013

Posted 20 February 2020 - 08:33 PM

Two points:

 

1) There is acquisition software that will support this. Voyager is one example.

 

Voyager Use Subframe 20200220.PNG

 

2) I personally would not use N.I.N.A. to control a remote observatory. I would want something with better condition handling for various off nominal events. Again Voyager is a fine choice if you have equipment that is supported. Alternatively, since it is an integrator, you can use equipment supported by one of the integrated packages such as Maxim DL or TheSkyX.



#9 james7ca

james7ca

    Cosmos

  • *****
  • Posts: 8,009
  • Joined: 21 May 2011
  • Loc: San Diego, CA

Posted 21 February 2020 - 03:37 AM

One issue, having too small of a field of view may leave you with too few usable stars to register your subs. That said, I've used small sensors (like the IMX178) for several years to do DSO imaging and I almost never had problems registering my subs. However, now that I'm using the larger IMX183 it's much easier to find stars for registration even when using very short exposures (like one second, sometimes less when doing so-called lucky imaging).

 

Also, having a larger field (always, without ROI restrictions) can provide more options for your final composition. Plus, you don't have to worry as much about getting (or returning to) an exact centering on the target.

 

I'm not saying that ROIs don't have uses in DSO imaging but there may be limits that you don't want to exceed, for cases as discussed above.




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