A mosaic question

[swag72]

I am looking at starting a Cygnus mosaic, but I have a question while I plan it and wait for the clouds to go!!

Do I need to take the same number of subs at the same exposure time for each panel?

[johnrt]

I am looking at starting a Cygnus mosaic, but I have a question while I plan it and wait for the clouds to go!!

Do I need to take the same number of subs at the same exposure time for each panel?

It would definitely help when it came to stitching them together. The closer you can match the frames the easier your life will be when putting it together.

[Psychobilly]

If at all possible yes...

[swag72]

Thanks both - Any other tips?

[swag72]

OK, here's a question - When doing just the one shot, I was looking at using my 24mm lens. If I am looking at a mosaic, would you use another lens instead? Perhaps a 50mm?

I want to use a prime so that I know I always have the same focal length.

[Psychobilly]

I used to mosaic using a 50mm f 1.8 (at f4) with zooms it tape them with micropore if it's likely to be a multinight project...

[jgs001]

I'd use whatever worked to fit the target in, and the number of frames you wanted to use for the mosaic... I came across this today from twitter, might be worth a look

Montage - Image Mosaic Software for Astronomers

[Psychobilly]

Definitely worth a look as you can output FITS from DSS...

[Catanonia]

I am doing the exact project and it definitely helps to keep exposures the same length and number.

I made the mistake of trying to add in ngc7000 with double exposure time to other parts and had awful times trying to stitch and match.

Also plan out and number your panes and stitch them together each night to ensure you are progressing right.

Remember to leave a good overlap to help your mosaic tool work.

Good luck, see my super mosaic in the image section As I am doing this right now and I am 11 panes of 300mm in.

[ollypenrice]

Being consistent in capture helps a lot but even then differences between nights affect the signal. Using Registar means that changes in focal length can be accommodated. If not using Registar, use primes or follow Peter above! If using Registar, try creating (or pinching!) a widefield as a template. Don't register each panel to another, register each to the widefield template and then combine them. You will have an easier time with large scale field curvature.

Olly

[Catanonia]

Being consistent in capture helps a lot but even then differences between nights affect the signal. Using Registar means that changes in focal length can be accommodated. If not using Registar, use primes or follow Peter above! If using Registar, try creating (or pinching!) a widefield as a template. Don't register each panel to another, register each to the widefield template and then combine them. You will have an easier time with large scale field curvature.

Olly

Is field curvature what is happening to my mosaic Olly in DSO ?

Noticing then even though I have not changed the image train, the images are steadily changing angle. Either that or my train has slipped.

[ollypenrice]

Is field curvature what is happening to my mosaic Olly in DSO ?

Noticing then even though I have not changed the image train, the images are steadily changing angle. Either that or my train has slipped.

That's what will be happening, yes. In my 9 panel of Heart and Soul to Double Cluster I had room to spare when I modelled the full image on the planetarium. When I put it together, though, the outer images curved away, creating gaps like a partly peeled orange if you like. When I cropped a rectangle out of it to avoid the gaps the Soul was a only just in! It had me worried... If not using a template the distortions will vary depending on the order in which you build the mosaic. Start with images from the middle and work out.

I was talking to Karel Teuwen about this and he said his giant Cygnus Loop was done by registering to a template image instead. If you register to the wide template image, however, that image form is what you will get on your mosaic.

Olly

[martin_h]

I was talking to Karel Teuwen about this and he said his giant Cygnus Loop was done by registering to a template image instead. If you register to the wide template image, however, that image form is what you will get on your mosaic.

Olly

Thats what I did for my Cygnus mosaic.....started from the middle and worked out from there, registering to a JPEG I found on the net.

[antimorris]

Yup - just to echo everyone else - it makes it SOO much easier when it comes to stitching if they are the same. My current mosaic project is 16 panels and still going and I am shooting them all at 10x1200s. Hardly had any stitching problems at all, which has been nice.

[swag72]

So as I understand this, it will be better to keep the focal length and subs the same. So I will start in the middle by Sadr and then work my way out and round. Will this cause curvature or will I be OK? I'm not sure what Olly suggests about a widefield template.

[themos]

There is no way of rendering a curved field on a flat plane without distortions. If you want a really wide field, you will have to view it with a panorama viewer like this one :

pan0.net :: Intro

[martin_h]

So as I understand this, it will be better to keep the focal length and subs the same. So I will start in the middle by Sadr and then work my way out and round. Will this cause curvature or will I be OK? I'm not sure what Olly suggests about a widefield template.

Using a widefield template - using Registar, will 'bend' the seperate images onto a flat background....dont forget you are imaging the whole by moving the scope from a fixed point in an arc to cover the sky.

[MikeD]

Same here, same sub lengths and as near as can be ( given sky conditions) same total imaging time per image.

I didn't use a template and have had no field curvature issues but that's probably because my mosaic is 1200mm @ F4.8.

Mike.

[swag72]

I didn't use a template and have had no field curvature issues but that's probably because my mosaic is 1200mm @ F4.8.

How does focal length and aperture affect field curvature? If I am imaging a mosaic at 50mm at F2.8 will I get curvature? How do you work this out?

[ollypenrice]

It will surely be down to how much sky you are covering. If you have a small chip and a long fl then you cover little sky so experience less curvature. Big chip and short FL covers lots of sky, so the reverse is true. The flatness of your optical field will also play a role.

Remember, the earth is locally flat. It starts to become spherical when you take a macro view. (Easy to pick holes in that but I'm sure you know what I mean.)

Olly

[Psychobilly]

Does the aperture setting come into it with fixed FL camera lenses? Varying the aperture will help with star shape but not alter the FOV... with scopes we normally change the FR by using Barlow's or reducers... keeping the aperture constant...

[themos]

How does focal length and aperture affect field curvature?

It's not field curvature we are worried about here (that only affects star quality). It's "distrortion" we care about, a geometrical transformation. That's what makes stitching panels together difficult.

I think up to 60 degrees of view is tolerable as the distortion will not be too much but that's only a ballpark figure that applies to monitor sized image or similar. If you make a 8000x6000 mosaic and view it at 100%, you will notice problems.

I suggest you start reading up on this software, I found it very powerful

Hugin - Panorama photo stitcher

[Catanonia]

ah ok got ya.

With my super mosiac of Cygnus I centered on the NAN and worked outwards. Also helps I am at 300mm as well.

[swag72]

Hugin - Panorama photo stitcher

Have you used this software for astro mosaics?

[ollypenrice]

It's not field curvature we are worried about here (that only affects star quality). It's "distrortion" we care about, a geometrical transformation. That's what makes stitching panels together difficult.

I think up to 60 degrees of view is tolerable as the distortion will not be too much but that's only a ballpark figure that applies to monitor sized image or similar. If you make a 8000x6000 mosaic and view it at 100%, you will notice problems.

I suggest you start reading up on this software, I found it very powerful

Hugin - Panorama photo stitcher

Yes, you are right, we should call it distortion. I come back to the idea of basing the mosaic on a widefield template, though. Isn't that going to be the most predictable way to get a result?

You only need s single sub from a camera lens to make one that will do the job.

Olly