cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Thu Jan 26, 2012 1:50 pm

I am looking at building a stand alone device for controlling a telescope mount, camera, focuser, etc etc.  I could easily do all that with an arduino OTHER than the guide camera bit which is basically a device that looks at a video signal of a star and moves the scope by minuscule amounts in order to keep that star in the centre of the field.  Normally this requires a computer/laptop sat next to the scope but you can get some standalone devices costing £400+ Having something with the processing power to assess the video input would make this a reality.

Shame they are not yet available.

PaulBuxton
Posts: 57
Joined: Tue Jan 10, 2012 11:38 am

Re: Astro photography

Thu Jan 26, 2012 2:04 pm

I thought that the normal technique for tracking a star with a telescope was to have a properly aligned equatorial mount and then you just need a motor turning at a fixed rate to track the star.

Tracking the star using video seems like a very awkward way of doing it, as the problem you will have is that one star looks very much like another, and at the light levels you have the noise on a digital camera will be quite high (unless you are using fancy cooled sensors or lots of image stacking).

Might it not be worth having a way of tracking the star based on it's known position? E.g. if the Pi knows where a star should be it can move the telescope to keep it in view. It would presumably need some calibration by pointing at a couple of known stars. But then you can have it point at any star you could name.

Just a thought.

Paul.

cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Thu Jan 26, 2012 2:10 pm

It is if you only want to take a 10 second shot of the moon but it you want to take 100 x 20 second shots of deep space in order to get high quality images of a nebula or distant galaxy then you need the scope to be held even more still than a high quality mount can manage.

Using a second scope sat on the same mount and having a video camera in it is common practice.

The video came is pointed at one specific star which the guide computer effectively keeps in the middle of its cross hairs.

PaulBuxton
Posts: 57
Joined: Tue Jan 10, 2012 11:38 am

Re: Astro photography

Thu Jan 26, 2012 2:33 pm

I pretty much exhausted my knowledge of astronomy in the previous post so can't argue with your knowledge of how these things work.

However to my mind there is a fundamental flaw in the video tracking method, in that it requires the star to have moved within the field of view in order to be able to correct it. Which means that your captured image will always have some motion blur introduced.

It is certainly an interesting problem!

You might consider having a look at OpenCV which has a good selection of image processing functions, including image tracking. I beleive that most USB web-cams will have some degree of support in the Fedora linux distro (certainly one of the first things I will be playing with will be using one I hope!) so are pretty easy to use with OpenCV (which I think will pretty much just compile easily enough as well).

Paul.

Joules
Posts: 87
Joined: Sun Dec 25, 2011 8:50 pm

Re: Astro photography

Thu Jan 26, 2012 2:39 pm

I used to have an SBIG ST-4 that was used for tracking, and could be used for imaging many many moons ago...  Yep, can see the Raspi being a good candidate for this application especially with a home build cooled webcam.

cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Thu Jan 26, 2012 2:42 pm

Remember that these things are tiny!  The movement of LESS than one pixel for the hole star can be detected by interpretation of the shape of the star across multiple pixels.  It does work and it is the basis of all long exposure astronomy photography.

I could easily build the entire device using an arduino but the video tracking part requires something a tad more powerful and I think that this is going to be it

I'm not an expert with the astro photography myself it is something I am just starting to look at but I love hobbies where I can mix bits of other hobbies

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 23636
Joined: Sat Jul 30, 2011 7:41 pm

Re: Astro photography

Thu Jan 26, 2012 2:45 pm

cowasaki said:


It is if you only want to take a 10 second shot of the moon but it you want to take 100 x 20 second shots of deep space in order to get high quality images of a nebula or distant galaxy then you need the scope to be held even more still than a high quality mount can manage.

Using a second scope sat on the same mount and having a video camera in it is common practice.

The video came is pointed at one specific star which the guide computer effectively keeps in the middle of its cross hairs.


I would have thought the processing would be better off at the image stacking stage, so adjust the images to match each other, rather than trying to keep the camera that steady. Or is the steadiness required for the 20s part rather than the 100 part? (Celestron C8 + EQ goto mount, no astro photography though)
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

User avatar
Gert van Loo
Posts: 2484
Joined: Tue Aug 02, 2011 7:27 am
Contact: Website

Re: Astro photography

Thu Jan 26, 2012 2:47 pm

All I can think of is that you could use it to first track a known star, use that to calibrate and extrapolate your system movement, then start making pictures. Or are there simpler ways to calibrate your system?

p.s. We do not yet have Camera boards which connect the the CSI port. You have to wait a couple of months for that to come.

cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Thu Jan 26, 2012 2:52 pm

Sorry I'll explain…

The camera is attached to a second smaller scope which is attached to the same mount as the MAIN SCOPE.  The main scope is moved to point at the primary target and then whilst it is still pointing at this primary target the second "guide" scope is moved slightly so that it points at ANY available star.  The computer "watches" the guide scopes output and it the star starts to wonder it sends a signal to the mount (which obviously has both scopes on it) telling the mount to move its motors to get the star back into the centre of the guide scope's view.

This way the main scope will effectively be locked at the same place in space.

It cannot be done afterwards because after 100 x 20 second exposures etc the target might be miles of the scope and even if it wasn't you want to use the whole screen for your shot.

This setup works

PaulBuxton
Posts: 57
Joined: Tue Jan 10, 2012 11:38 am

Re: Astro photography

Thu Jan 26, 2012 3:40 pm

cowasaki said:


I could easily build the entire device using an arduino but the video tracking part requires something a tad more powerful and I think that this is going to be it


Definitely recommend checking out the OpenCV libraries then.

Another thing you might consider is keeping the camera static during your image exposure, then after the exposure is complete use the tracking to reposition the telescope before the next exposure.

To correct the streaked line you would use Image Deconvolution, which (given an accurate describtion of how the image is blurred) can do an amazing job of correcting it.

Paul.

cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Thu Jan 26, 2012 3:53 pm

PaulBuxton said:


cowasaki said:


I could easily build the entire device using an arduino but the video tracking part requires something a tad more powerful and I think that this is going to be it


Definitely recommend checking out the OpenCV libraries then.

Another thing you might consider is keeping the camera static during your image exposure, then after the exposure is complete use the tracking to reposition the telescope before the next exposure.

To correct the streaked line you would use Image Deconvolution, which (given an accurate describtion of how the image is blurred) can do an amazing job of correcting it.

Paul.


I appreciate what you are saying but the reality is that what I"ve described is the best way that has been found.  These mounts are designed to take the input nudge this way, nudge that way etc signal from a guide scope computer.  Also if you are taking a picture of something that is filling 90% of your screen you have to keep correcting the position because if you just left it then part of the image would escape the bounds of the screen quite quickly.

Imagine looking at the side of a coin a mile away and blowing the image up till it fit the whole screen. The slightest little movement is going to have part of the coin outside of the screen.  The mounts a phenomenally good but that tiny little tweak is why a guide scope is used.

I will definitely look at the software that has been spoken about.  There is already linux software available including source to accomplish this task so I should hopefully be able to port it and once ported I can add the extra stuff quite easily.

Thanks

PaulBuxton
Posts: 57
Joined: Tue Jan 10, 2012 11:38 am

Re: Astro photography

Thu Jan 26, 2012 4:10 pm

I didn't realise that the image would be filling that much of the view. I guess in that case you have no choice but to track it.

Have fun!

User avatar
rick_2k
Posts: 8
Joined: Thu Jan 26, 2012 4:16 pm
Location: South East, UK
Contact: Website

Re: Astro photography

Thu Jan 26, 2012 4:22 pm

Will be using my pi for astro use

Sounds like its more than capable to be a good tracker and / or multifunctional device maybe even handling image capture.

Move > acquire image > move etc and then store frames safely for stacking or whatever later on!

Who knows!

Cant wait.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 23636
Joined: Sat Jul 30, 2011 7:41 pm

Re: Astro photography

Thu Jan 26, 2012 4:25 pm

Should have said in my post about using the processing during stacking to realign, that I assume the pictures were fairly well aligned already as they would be taken on a tracking/goto mount.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

User avatar
perfo
Posts: 92
Joined: Mon Jan 30, 2012 2:42 am

Re: Astro photography

Mon Jan 30, 2012 3:26 am

Showing my limited knowledge here but don’t some webcams
have face tracking software built in? Not sure it will be of any use but if you
could hack your way into this signal then it could be used as auto star
tracking instead. Maybe it will think of a star as a face already and thus
making that bit a little easier.

User avatar
Redrobes
Posts: 80
Joined: Mon Dec 26, 2011 9:19 pm
Location: S.W. UK
Contact: Website

Re: Astro photography

Mon Jan 30, 2012 1:36 pm

Isn't the normal method to use a quadrature photodiode and drive the tracking motors from the error off of the 4 photodiodes instead of using a full CCD. Whilst I agree that this does mean that you need the tracking star in the middle of the tracking scope, it does mean that you can continuously monitor it instead of doing frame capture, analysis and then tracking error generation.

You can get a microcontroller like the Arduino to run this for about 10mW instead of the 5W your gonna need with a full blown linux PC. You could run that off of a couple of AA batteries or a PP3 instead of more like a small lead acid or RC battery pack.

Whats the advantage of doing it all with a CCD + PC ? After all your just trying to use a few pixels of the CCD as photo diodes and ignoring all of the rest anyway. Is it just the convenience of being able to pick the star off of the LCD display to make the line up easier ?

http://pinout-circuits-images......B-C203.jpg

Joules
Posts: 87
Joined: Sun Dec 25, 2011 8:50 pm

Re: Astro photography

Mon Jan 30, 2012 2:33 pm

You can track and image from a CCD, a smart system could estimate its required track from a short pre alignment run, then allowing long exposure and check it"s tracking on completion of each image prior to stacking.

cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Mon Jan 30, 2012 3:43 pm

Redrobes said:


Isn't the normal method to use a quadrature photodiode and drive the tracking motors from the error off of the 4 photodiodes instead of using a full CCD. Whilst I agree that this does mean that you need the tracking star in the middle of the tracking scope, it does mean that you can continuously monitor it instead of doing frame capture, analysis and then tracking error generation.

You can get a microcontroller like the Arduino to run this for about 10mW instead of the 5W your gonna need with a full blown linux PC. You could run that off of a couple of AA batteries or a PP3 instead of more like a small lead acid or RC battery pack.

Whats the advantage of doing it all with a CCD + PC ? After all your just trying to use a few pixels of the CCD as photo diodes and ignoring all of the rest anyway. Is it just the convenience of being able to pick the star off of the LCD display to make the line up easier ?

http://pinout-circuits-images......B-C203.jpg


I will look into that.  If I can get an input saying left, left, up, left, down etc then I could knock the whole lot together in a few days.

Dazzler
Posts: 14
Joined: Mon Jan 09, 2012 6:26 pm

Re: Astro photography

Mon Jan 30, 2012 4:13 pm

Cowasaki - is that Darren from TP?

cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Mon Jan 30, 2012 4:54 pm

Dazzler said:


Cowasaki - is that Darren from TP?


It certainly is

cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Mon Jan 30, 2012 5:31 pm


Redrobes said:

Whats the advantage of doing it all with a CCD + PC ? After all your just trying to use a few pixels of the CCD as photo diodes and ignoring all of the rest anyway. Is it just the convenience of being able to pick the star off of the LCD display to make the line up easier ?


It does tend to be done that way with cameras such as the QHY5 being favourites.  A matrix of pixels would be ideal although being able to look through the scope and have the detector would be ideal.  This would be in order to point the guide scope at an object……  Still thinking about it though…..

timgiles
Posts: 101
Joined: Thu Jan 12, 2012 8:58 am

Re: Astro photography

Mon Jan 30, 2012 6:03 pm

Hi - I posted to Bit-Techs preview of the Raspberry Pi my intention with the second unit I purchase (probably in 4 to 6 months time) to develop the RPi as an improved ultra low power guider.

Whilst yes, we can all hope for a 'perfectly' aligned scope, those with Alt Az mounts (mounts that go 'left/right' and 'up/down' on a level plane) can be controlled with a guide scope (read RPi with webcam) for medium term exposures. You still have the problem of sky/field rotation, but that can be sorted out later and is only a problem with long exposures. Obviously with a well aligned EQ mount, it is a lot easier to guide for longer periods.

Also I wonder for those who have domes, whether the RPi and a gert board or similar would allow for a superior dome controller. You could monitor everything, control a huge amount, send instructions to the RPi remotely to control the imaging. Possibly even tie in the security and weather facilities depending on how much your observatory has.

Pie in the sky for the moment, but certainly I look forward to trying something with a second RPi that is sky related.

cowasaki
Posts: 16
Joined: Tue Jan 10, 2012 5:55 am

Re: Astro photography

Mon Jan 30, 2012 7:19 pm

timgiles said:


Hi – I posted to Bit-Techs preview of the Raspberry Pi my intention with the second unit I purchase (probably in 4 to 6 months time) to develop the RPi as an improved ultra low power guider.

Whilst yes, we can all hope for a 'perfectly' aligned scope, those with Alt Az mounts (mounts that go 'left/right' and 'up/down' on a level plane) can be controlled with a guide scope (read RPi with webcam) for medium term exposures. You still have the problem of sky/field rotation, but that can be sorted out later and is only a problem with long exposures. Obviously with a well aligned EQ mount, it is a lot easier to guide for longer periods.

Also I wonder for those who have domes, whether the RPi and a gert board or similar would allow for a superior dome controller. You could monitor everything, control a huge amount, send instructions to the RPi remotely to control the imaging. Possibly even tie in the security and weather facilities depending on how much your observatory has.

Pie in the sky for the moment, but certainly I look forward to trying something with a second RPi that is sky related.


Tim,

Sounds like someone with a similar interest  I run an NEQ6 Synscan and have just ordered an ST80 which should arrive tomorrow to use as a guide scope.  Although I was going to buy a standalone guide scope controller (to begin with) I would love to build a device with all the astro functions built in.  My plan was to use a peli case with everything built into it.  Guiding, camera control, focus control, sky quality meter, control, gps etc but then maybe other stuff as we go along   My original plan was to take a laptop apart and fit it inside WITH the arduino stuff etc BUT this might be a more interesting idea even if I need to use two of them

timgiles
Posts: 101
Joined: Thu Jan 12, 2012 8:58 am

Re: Astro photography

Tue Jan 31, 2012 11:32 am

Yes - my current server (read. laptop locked in a cupboard) that looks after web, file and ftp, is also in the offing as I replace it with a Synology NAS next month. After the move to north Sweden last year, I cant believe how much clearer the night sky is. Northern lights are great at the moment.

As I mentioned, I wont be starting my project till the summer, but I will message you once I do to let you know how I get on. In terms of kit, I had to sell most of mine before coming over and am looking at buy a new mount in 2 or 3 months time. Scope wise, my 8inch LX200R was sold as well but I am using my William Optics 70mm Zennithstar (CF tube) and two solarscopes, the PST and PST Cak.

Another 'project' for the summer, I am planning to try and write a program that will read from two web cams, so i can use both the PST and PST CaK at the same time, and be able to have the images either overlaid or switch between them to see the sunspots 'rising and falling' between the temperature layers.

Ahhh, just need TIME. Of course, with 24hr sunlight during the summer, helps a bit too!

peter.polakovic
Posts: 3
Joined: Wed Aug 22, 2012 6:57 pm
Contact: Website

Re: Astro photography

Fri Aug 24, 2012 6:10 pm

Raspberry is very promising hardware for standalone or remote telescope control. It is quite easy to make INDI running on it and I tried simulators, QHY5 and two different SX CCD cameras with no problem at all (at least with jINDI as a remote client).

The only missing piece for perfect guider is connecting (and powering) Raspberry directly to (and from) EQ6 mount but some 3.3V to 5V logic level converter (and 12V to 5V) converter will be necessary. Is there some hardware guru to try it? :)

Image
Image

Return to “Other projects”