Registering and stacking - problems encountered

[tnkumar]

I took 8 exposures of Alnitak, 1 sec each EQ mount not polar aligned and no tracking.
Images were captured as raw files.

I am not able to stack these.
The files need a sharp stretch to bring up the details of other stars but the background is very noisy after stretching.

Any idea what I might be doing wrong that prevented stacking?

I can upload my raw files if needed.

[tnkumar]

I uploaded one of the raw files https://skydrive.live.com/#cid=DB19065F7A789715&id=DB19065F7A789715!103

[Nocturnal]

Without seeing the console log we can only guess as to what's going on but 1s exposures are close to worthless, sorry. Not really worth the trouble stacking those or figuring out why it's not working.

How about polar aligning your mount and taking some 30s exposures so there will be some stars visible?

A stack is at least two images. If you have two images you think should register OK but don't, upload them both.

[tnkumar]

Second one also uploaded

[tnkumar]

I am also attaching stretched .jpg files of the two images with this e-mail

[tnkumar]

Should we do registration and stacking before stretching or after?

[Josh Lake]

The basic order is:

1. Calibrate (if you are using darks, bias frames, or flat fields, though it looks like you are not)

2. Register

3. Stack (Image Integration)

This will give you a single frame (or a frame per color channel). Then you should do several things to that frame before doing any kind of permanent screen stretch. But you can use the Screen Transfer Function to see your data as much as you want -- it does not affect the linear data, just displays it differently.

[tnkumar]

I believe that in my images, there is a lot of background noise that is causing problems with registration.
When I did a stretch, I actually clipped many of the nose pixels.
By doing so, I felt that the program may be able to register more easily.

Is the amount of noise in my image something normal?

I guess there are two ways this background noise can be removed
- longer exposure times
- taking dark frames and adjusting for the noise

Any other suggestions?

[zvrastil]

Hello,

the problem of your images is not high amount of noise, but signal-to-noise ratio. In other words, noise is ok, but your signal is very low and is comparable to noise. That's the reason why you see so much noise after stretch. Subtracting dark won't help - it's used to subtract dark current which is zero for such sort exposure. It will rather make things worse. The only reasonable way is to make longer exposures to improve signal-to-noise ratio.

regards, Zbynek

[tnkumar]

Thanks all for your inputs.
I will spend time to get my scope polar aligned and then take a longer exposure.

One question though - upto what exposure times are ok for non polar aligned images?

[zvrastil]

One question though - upto what exposure times are ok for non polar aligned images?

That's hard to say as it highly depends on scope focal length, quality of the mount and of the polar alignment. Simply several exposures with different times and check the shape of the stars - as long as they stay round, you're fine.

[georg.viehoever]

If you use a stationary camera (no tracking): for 85mm, it is 10 seconds. for 850mm, it is 1 second, etc.

[tnkumar]

Thanks all. These replies help.
My SV NightHawk 80ED is 560mm. So, approximately, even a 1.5 second exposure would work.
With 1 sec, I got round stars.

I believe the best thing for me would be to polar align, enable tracking and take longer exposures - 30 secs and also get some dark frames.

This should enable me to get good quality images to work with.

[georg.viehoever]

...I believe the best thing for me would be to polar align, enable tracking and take longer exposures - 30 secs and also get some dark frames.

You are right, this is the way to go. Note that you can do some fairly nice wide fields without tracking, see for example http://pixinsight.com/forum/index.php?topic=2068.msg13358#msg13358. Fo 550mm focal length, tracking is mandatory: Exposure times around 1 sec are just not enough to see more than with your eye. To get a feeling about the acceptable exposure times, see http://www.gabrielditu.com/astronomy/astrophoto_calculator.asp

Georg

[tnkumar]

It is nice to see what beautiful pictures are possible with no polar alignment and no tracking. Thanks for sharing the link.