The Astro-Imaging Thread

[YNM]

Image noises can be dealt with by using some program that removes them. But they're going to ask you for two (or three) more images :

- Dark images (capture while closing the whole optical pathway to darkness, to the same exposure time). Used to reduce random noises from your object image ("hot pixels").

- Flat images (using a uniform screen, can also be done by pointing at cloud covers over cities or white wall, or by some device called a flat box, take the same exposure length - well in case it'd overload your detector use a shorter length). Used to calibrate the relative sensitivity of all the detector pixels. This is done by "dividing" the object image by the cleaned flat images (flat minus dark flat images).

- (optional) Dark Flat images (dark images for your flat takes in case it's of a different exposure length). Used to subtract random noises from your flat images.

Most softwares, given the required files, will do the job automatically.

In case it's still noisy, editing could be your other friend. Or it's just a suboptimal night.

 

Edited January 1, 2017 by YNM

[Jonfliesgoats]

My ambitions may exceed my skills and equipment for now.  I was thinking of ways to keep kids excited about the cosmos (see my thread regarding pessimism in science for my motivations there.).  Having a special trip to a nearby dark sky location and peering at the moon through our telescope may do the trick.

[YNM]

Jupiter shall do nicely - lots of details to be looked on for the king.

Also, the moon is currently close to the Sun...

[Jonfliesgoats]

Good idea with Jupiter!

[Green Baron]

9 hours ago, ProtoJeb21 said:

Also, a quick question. I really want to get rid of the image noise in the star cluster photos. However, if I make copies and stack them, it won't work. Instead I'm thinking of another method: make copies, but tweak every image so that its mostly like the edited one shown above. That way each image used for compression is slightly different and won't result in something so close to the original. Would this work? (Tagging @Green Baron)

Hi @ProtoJeb21,

first the vignetting, that's the dark edges. Source is either the scope's objective/mirror that has an image field that is too small for the size of your cameras chip (or the chip is too large :-)) or something in between (focuser, corrector, filter element, ...) is narrowing the way. If you can live with it i suggest you do it for now. Later on with more proficiency, as @YNM says, flatfield frames and the corresponding dark frames could partly help. Though the vignetting is probably too strong to be corrected.

I kindly ask to search the web for the meaning and sense of darkframes in general, some theory cannot harm :-) This leads us to why astronomic cameras are cooled. This is, in fact, not only to reduce noise, but mainly for reproducibility. And why we always use raw files from the camera, never any jpegs or tiffs it might produce.

Switch off any automatism of your camera, use manual mode and raw files.

Flat fields are made on site, with the fully assembled and adjusted equipment, the scope covered with a grey blanket or, better a flat field foil in a rigid frame. Flat fields, as the name says, are representative for varying illumination that originates from the setup or maybe nearby external light sources. Like vignetting, reflections in the tube or focuser, dust on filter surfaces, ... Exposure time, camera iso setting and chip temperature don't matter but must be all the same, it should be long enough to get an evenly exposed picture without any saturated or grossly underexposed pixels. The corresponding dark frames must, like any dark frame, be made with the same exposure time, iso setting and chip temp. as the flat field frames it belongs to ! You can take the camera off the scope and cover it with the lid, but do not warm it.

As to the number of flat field and dark flat field frames: like with everything many are better are better than few, for practical reasons let's say 16 each are sufficient. Btw., only few people do flat field frames. The equipment these days is of good enough quality so that the sources of uneven illumination can be eliminated before they arise. And it's a lot of hassle, more than keeping stuff clean and connecting everything safely and securely. That's why i suggested to skip it.

 

The next is the noise, and there something can be done and must be done because that will reappear with every night shift. Copying a file is not involved anywhere, except for backups later :-)

Make your light frames with the setup (i assume a dslr and a tracking mount), a certain iso setting, exposure time and eventually aperture if using a camera lens. Do not switch ! At least 10 should be ok, more are better. Use a timer if your dslr doesn't allow exposures times longer than 30 seconds, like mine. If you don't have a tracking mount then experiment with exposure time, but if you use a telelens you'll have prolonged stars after seconds.

Then (or in between, if you dare to touch without moving anything), make your light dark frames (the dark frames for the light frames, which are most certainly different than the dark frames for the above flat field frames) with the same camera settings and temperature as above. Make 16 (more is better but impractical). So, if you have made 10 light frames, each 2min, 800iso, then do 16 dark frames, 2min, 800iso. You can take the camera off and cover it with the lid to avoid external light (protect the open focuser with a plastic bag), but do not warm the camera.

The next are the bias frames. These are easy. Chip temperature does not matter, but iso speed does and time. A bias frame contains the glowing of the electronics at the iso setting. Ideally it has no chip information but that's not possible with a dslr. So just take the shortest possible exposure time and do at least 16 frames.

 

Ok, to summarize: >10 lightframes, ~16 darkframes at the exactly same settings and conditions than the lightframes, ~16 bias frames at the shortest time and same iso setting as the lightframes organized as raw files on your harddrive in a matter that you do not mix them.

Feed DeepSkyStacker with them ...

If you still have a noisy background that is most likely from light pollution, which is not covered by the process, it fact it cannot be covered by any process known to me.

 

I was not successful last night, i had to wait until after midnight and then a high veil of clouds came up.

@_Augustus_: beautiful telescope !

 

1 hour ago, Jonfliesgoats said:

My ambitions may exceed my skills and equipment for now.  I was thinking of ways to keep kids excited about the cosmos (see my thread regarding pessimism in science for my motivations there.).  Having a special trip to a nearby dark sky location and peering at the moon through our telescope may do the trick.

 

Suggestions: Andromeda galaxy, open clusters, (young stuff for the kids), globular clusters (old stuff for the grown ups), Orion nebular, Venus in the evening sky (half venus on the 17th ?), a hike along the moons terminator line, structures in the grazing light are a fine sight.

 

[Wjolcz]

Today's Moon. Considering this was taken with my Astromaster 90AZ and 8 Mpix Sony Xperia L I'm pretty happy with the results:

Edited January 8, 2017 by Veeltch

[kurja]

On 20.9.2016 at 7:41 AM, MaxPeck said:

@Diche Bach

Stacking copies of one picture is not helpful as the whole point of it is to avarage out random variations.

I find dark frames are quite useless with dslr cameras, as raw from them isn't genuinely raw; calibration doesn't quite work like it should.

Light pollution does not increase noise, it usually just introduces a gradient or drowns out your signal. Calibrating your lights does not help, but there are processing options; search for gradient removal and/or background extraction. Some filters may be of help when taking the lights.

 

(sorry about the broken quote, stupid phone...)

Edited January 8, 2017 by kurja

[kurja]

On 12/31/2016 at 5:49 PM, _Augustus_ said:

No. GoTo scopes can't even find satellites.

What scope?

I'm not so sure, a goto mount will point which ever way you tell it to - why not at a satellite?

 

https://www.youtube.com/watch?v=Mi9ZOiaLKxc

[MaxPeck]

28 minutes ago, kurja said:

I'm not so sure, a goto mount will point which ever way you tell it to - why not at a satellite?

 

https://www.youtube.com/watch?v=Mi9ZOiaLKxc

I've got an iOptron CEM25 and I'm pretty sure it'll point at the ISS.  I think as long as you have a mount with with RA and DEC drives, you can pretty much track anything you want.  Mine'll track planets and comets as well.  I'm a member of a few Astrophotography clubs on Facebook and people routinely post pictures of the ISS where you can easily resolve the solar panel arrays and main structure.

On 1/8/2017 at 4:59 PM, kurja said:

Stacking copies of one picture is not helpful as the whole point of it is to avarage out random variations.

I find dark frames are quite useless with dslr cameras, as raw from them isn't genuinely raw; calibration doesn't quite work like it should.

Light pollution does not increase noise, it usually just introduces a gradient or drowns out your signal. Calibrating your lights does not help, but there are processing options; search for gradient removal and/or background extraction. Some filters may be of help when taking the lights.

 

(sorry about the broken quote, stupid phone...)

I use a DSLR and I always shoot dark frames.  Yes, the camera does some post-processing on the images, but it does the same processing on subs as it does on darks, so there's still some sensor noise to deal with.  I always shoot darks and flats and bias frames.  I also use Sequence Generator Pro (a great piece of software, I highly recommend it) which saves images as FITS files instead of jpegs, so that helps preserve some detail as well.

[kurja]

8 minutes ago, MaxPeck said:

I use a DSLR and I always shoot dark frames.  Yes, the camera does some post-processing on the images, but it does the same processing on subs as it does on darks, so there's still some sensor noise to deal with.  I always shoot darks and flats and bias frames.  I also use Sequence Generator Pro (a great piece of software, I highly recommend it) which saves images as FITS files instead of jpegs, so that helps preserve some detail as well.

Have you compared a same set of lights with and without darks? I never saw any benefit so I just don't bother with darks, I just use flats, and bias because flats don't come out right unless adjusted for bias. I use pixinsight, it can save in fits too, I usually save my "final iteration" in 16b tiff though.

My tracking is never spot on anyway, so as the images are aligned before integrating, bad pixels and such sensor problems get solved as a free bonus Maybe darks help if your tracking is pixel perfect, but who has that kind of tracking and shoots with a normal camera...

[Green Baron]

I did a direct comparison a little more than a year ago, when i startet with astrophotography. I recall the data but don't have the pictures any more. I tried with a starfield around Andromeda, good old Nikon D700, 800iso, 200mm Telelens, f/2.8, 2min and 4min exposures on a tracking mount. It was a warm night (maybe 22°C). I used DSS for stacking.

Without dark frames single shots were just noisy :-/. From 10 stacked exposures on things got better, but there was still visible noise. With 20 exposures the noise did not disturb me any more in a wide field photo. But i understand that for deep sky objects, where any post correction of noise destroys parts of the nebulosities, this is not recommended.

On the other hand 10 exposures with the subtraction of 10 dark frames and 10 bias frames really where an eye opener, i could hardly see any noise any more. I only shirk sitting double time on the hill, so I then decided to give myself the allowance to acquire a cooled ccd-cam.

And then the weather decided to show me what "natural advantage" means ... that was 8 weeks ago. And now the moon is shining so bright you can read a book outside ...

 

 

Edited January 10, 2017 by Green Baron

[kurja]

Ah, your warmer conditions may play a part here, I'm at >60 latitude so summer time is no-no for astrophoto and the temperatute on clear winter nights is pretty much always quite far in the negative degrees.

[Green Baron]

That may well be. And the cmos chips have more problems with noise than ccds. A single 8min shot, cooling inactive, with the ccd cam shows far less noise than a 1 min shot with the dslr.

 

Edited January 10, 2017 by Green Baron

[kurja]

Although if you're looking at single exposures, can you really tell if it's thermal noise, which appears in a static pattern and can be helped with darks, or if it's shot noise, which is random and can be helped by integrating more lights?

I haven't really seen what would clearly be thermal noise from any dslr cmos, but then again I only do astrophoto when it's cold outside. On another forum I saw a rather thorough test, result was that noise increases steeply after cmos sensor temperature exceeds 0C and it warms up by about 10C during continuous use... So -10C or below, and we should be good to go

[MaxPeck]

Honestly, yes, I can see a difference in my images once I start stretching the histogram.  Darks are absolutely essential.  I also auto guide and dither so that helps a bit too in reducing the noise and increasing the signal.  I tend to stick in the 2-3 minute exposure range because I'm in a Bortle 6 zone, and any longer starts getting light pollution washout.  I've done an 8-minute guided exposure on Traingulum and you can see a fair amount of detail, but trying to reduce the light pollution without losing nebulosity is a straight-up nightmare.

I prefer to save my subs in FITS format up until final processing.  I'll save my stacked and hacked version as a TIFF and then my publish-able version as a PNG.  I like to maintain the FITS format as long as possible though, so that I can retain the embedded header information in it - with SGP, you can plug in an image as a target and it'll read the FITS data and auto-slew the camera back to that spot so you can take more shots - makes accumulating data over multiple sessions a lot easier.  I'll save my first shot as a reference shot and then if I'm revisiting a target, I'll use that as my key to make sure the camera is pointed the same way.  SGP does astrometry plate solving automatically, so it's as simple as showing it what you want to shoot, letting it aim the camera and take some comparison shots to plate solve and then letting the rig get to work while I play Halo on the couch.  I'm okay with letting the mount and SGP do the heavy lifting on the front end, since I know I'm going to spend hours on the back end reviewing the subs, stacking, stretching and processing.

Edited January 13, 2017 by MaxPeck

[Green Baron]

I already asked myself what to do to meet the same field of view if one night is not enough time. I only will have to do it manually.

8 min are not that much of a problem here. The ccd cam (atik 383l monochrome) only offers fits or tiff as output formats and the docs say that fits (16bit integer) corresponds to raw data, so there is no choice for me.

I found out that i must stay near when pointing at objects high above because the camera could hit a leg of the mount ...

Also i prefer ksp or factorio when waiting :-)

 

[MaxPeck]

Getting a good CCD camera is on my someday list... I'm locked into the DSLR right now because I blew my last astronomy allowance on the CEM25.  I need to find some new revenue streams to offset the black holes of funding that are my kids before I can start buying new toys.  I'm jealous of your ATIK, I'd love to have a dedicated astronomy cam.  And a permanent pier.  And a dome.  And... and...

 

 

[_Augustus_]

Venus and part of the Moon on the 1st. Taken with a Logitech Quickcam 4000 and Celestron NexStar 4SE, processed in Registax.

 

[Wjolcz]

Nice. Why is it so purple? Filter or something?

I saw Venus about a month ago through the telescope but it was so cold holding a phone and taking the pics was near impossible (shaky hands). Now that I've made myself a phone-telescope adapter I'll try and take some maybe next week when the weather is nice.

EDIT: I lack the budget and it means there's a lot of experimenting involved in my astronomical adventures. Has any of you tried X-ray film as a filter used to take pictures of the Sun? I hope pointing my refractor telescope at it won't melt it or something?

Edited February 4, 2017 by Veeltch

[Green Baron]

8 hours ago, Veeltch said:

Nice. Why is it so purple? Filter or something?

I saw Venus about a month ago through the telescope but it was so cold holding a phone and taking the pics was near impossible (shaky hands). Now that I've made myself a phone-telescope adapter I'll try and take some maybe next week when the weather is nice.

EDIT: I lack the budget and it means there's a lot of experimenting involved in my astronomical adventures. Has any of you tried X-ray film as a filter used to take pictures of the Sun? I hope pointing my refractor telescope at it won't melt it or something?

Ages ago i have taken photographs with infrared films (other side of the spectrum :-)). Nice for landscapes. Roentgen radiation (x-ray) is a broad spectrum between uv and gamma radiation, much broader than visible light. You'd somehow have to convert it to a visible light spectrum and send it to a ccd. I can imagine that this is a challenge for an amateur, but i do not know.

As for film, it's expensive these days and the chemicals aren't always available anymore. You'd probably have to mix your own soups or buy expensive ready made stuff if available. If you think of medical equipment, they use a radiation source for exposure (cesium or something), a "bright shining light" sotosay. I can imagine (but i do not know for sure) that medical equipment is simply not sensitive enough. Also, with film you have one shot per negative that might cost at least a few funds per shot and development, and that's it. No delete and redo (F5/F9) like these days :-)

Sorry for shouting, it is not my usual habit, but IF YOU LOOK AT THE SUN ALWAYS USE A SOLAR FILTER ! You'll damage your eyes and your telescope from the heat. Solar filter foils are available at astronomy shops. Get a sheet, cut it out and fabricate a sturdy frame (plywood) to cover your front lens or opening so that the filter cannot fall off. Additionally, you can use projection on the other end on a white sheet of paper or so. I am sure you can find examples.

For solar photography people usually use narrow band filter, like hydrogen alpha e.g., but i fear these interference filters are extremely expensive (200 funds a version with just 48mm diameter), and before mastering narrow band maybe the "normal" rgb-techniques should be mastered, because for presentation it is all about combining the wavelengths to a nice and shiny colourfull picture :-) See astrobin for examples ...

 

Ok then, i'm looking forward to what a phone camera can deliver these days :-) You are right using adapters, it will make it a lot easier to find the focus and hold still. If you need help with that try bahtinov- / hartmann- /carey-mask that you can cut out as well, no need to spend money for that.

 

[Wjolcz]

@Green Baron I was actually trying to say that I don't want to take pictures in X-ray bands, but instead cut out small pieces out of the X-ray film with the picture of my hand bones. And I won't be trying to look through it myself. I'm no a madman (yet). I'll try and take a few pictures with my phone just to see how it will look. Also thanks for the tips.

Edited February 4, 2017 by Veeltch

[_Augustus_]

13 hours ago, Veeltch said:

Nice. Why is it so purple? Filter or something?

I saw Venus about a month ago through the telescope but it was so cold holding a phone and taking the pics was near impossible (shaky hands). Now that I've made myself a phone-telescope adapter I'll try and take some maybe next week when the weather is nice.

The problem with taking afocal shots of the planets with your phone is the exit pupil of your scope. The exit pupil needs to be around the size of your phone's camera, but at high power (which it needs to be at for the planet to take up a large portion of the frame) with a scope under 12" it's not going to be that big. Thus I hacked together this webcam.

I don't know why it's purple and green like that. I'm using a Mak. Probably an issue with the sensor.

 

1 hour ago, Veeltch said:

@Green Baron I was actually trying to say that I don't want to take pictures in X-ray bands, but instead cut out small pieces out of the X-ray film with the picture of my hand bones. And I won't be trying to look through it myself. I'm no a madman (yet). I'll try and take a few pictures with my phone just to see how it will look. Also thanks for the tips.

You'll still need a solar filter....

Edited February 4, 2017 by _Augustus_

[monophonic]

53 minutes ago, _Augustus_ said:

You'll still need a solar filter....

He is trying to ask how used X-ray film would work as solar filter with a phone camera.

[Wjolcz]

18 minutes ago, monophonic said:

He is trying to ask how used X-ray film would work as solar filter with a phone camera.

Man, my syntax is poor. But yeah, that's what I meant.

[MaxPeck]

1 hour ago, monophonic said:

He is trying to ask how used X-ray film would work as solar filter with a phone camera.

Just get some Baader solar film. It's not that expensive and much cheaper than replacing your phone, camera or eyes.