Big Tassie Vela with bonus gradients!!

[graham.hobart (Graham stevens)]

Hi All, I am still working through the mad week I did with three clear nights in a row and narrow band. I hope this is the final version of part of the Vela SN - it's a combination of two different works in Planning - but not enough to mosaic it.
I got 7.46 hrs Ha Baader and 6.41 O3 to make it .... 13.87hrs bicolour narrow band - but I used O3 in the green and blue as well to balance it.Ha red
I stacked sigma clip Maxim Dl - (Maxim guided as well), preliminary processing in Star tools, did a bit more in PS 5, also tried my first proper go at Pix- did a DBE trying to get rid of that gradient on the right hand side, not so sure I did any good.
anyway- I present you my last Vela for this year ( I hope)
Baby FSQ 85 FLI mono 8300 Baaders PMX...
Cheers and I hope you give me some good advice all....
Graham
NB The gradient comes from shooting from suburbs and stuff I think ? or is it just my horror flats?

[rustigsmed (Russell)]

hi graham,

it looks to me that it could be a flats issue for the blue and green channels only. have you tried stacking it all the channels together without applying flats to see how it turns out?

great field.

russ

[graham.hobart (Graham stevens)]

Hi Russ, no- but that's a good idea mate- will give it a go tonight

[gregbradley]

A good image there Graham. The gradient can be processed out.

One way is to apply HLVG filter (in PI its called ACDNR I think) to remove the excess green which light pollution usually is.

It will thow your nice teal green off so you'll have to rebalance.

There is a gradient technique using channels. I haven't used it in a while but in your case where the gradient is on one side only it would work really well.

You can PM me if interested for the procedure.

Greg.

[Atmos (Colin)]

Very nice nebula Graham, the Baby Q looks like it is performing really nicely! I personally largely haven't noticed much in the way of gradients with my 3nm narrowband filters from the burbs of Melbourne, light pollution is pretty much negated due to the narrow band width. I would guess flats are the issue then with a slight overcorrection on the RHS.

What are you using to do flats? It could be caused by an EL panel that is not quite flush or sky flats that are not on the opposite side of the sky to the rising/setting sun.

[Tony_ (Tony)]

Nice image Graham.

There is a free software called fitswork which is quite good for removing gradients when there is a lot of nebulosity present. You can remove the gradient from this in a few minutes without affecting the colours much.

Tony.

[Shiraz (Ray)]

Yep, it has gradients for sure, but it is still an very impressive image that somehow conveys an idea of the vastness of this feature. nice work.

[graham.hobart (Graham stevens)]

Thanks all- I will look into fitswork and Greg I will PM you about that technique if you don't mind
Cheers
Graham

[Rex]

Hi Graham,
Really nice detail in your image mate, and Loving the colours. If you have PI there is another process called dynamic background extraction which is quite good for removing gradients etc. I copied your image and ran DBE on it a few times and then adjusted the histogram, and had to resize again after that coz it went over the 200k limit for the site. I have attached it below. There is a heap of tutorials on DBE around and when you do it properly it can take a while and a bit of mucking around to make sure samples are all in the right spots, but for this exercise I just let it choose it's own sample positions and ran it a few times. Hope that helps mate. PI has so many processes it's really hard to get your head around all of them.

[graham.hobart (Graham stevens)]

Cheers Rex- I had a go before with PI but I realised I was subtracting the bg not dividing by it- will have another go prob this weekend

[gregbradley]

Gradient channel technique.

Open image in PS
click on channels
click on add new channel
call it gradient
click on the gradient tool select the type of gradient which shows white triangle in one corner the rest black
drag a line about 1/4 the way from the right side where the gradient is.
It will create a white patch that fades out to the distance you dragged.
now click on rgb in channels
click on load selection and select gradient
now the area you made as a white patch shows up as a dotted line about 1/4 to the left of the right hand side of the image. Now any tool adjusts only that area.
Use levels and reduce the brightness of the gradient affected area. In your case that is mostly in the green channel so do it generally but also select green in levels and reduce that. That really got rid of it.
Unselect when done and you are finished.

Really you could achieve the same result more simply by simply lassooing the affected area then blurring the edge under refine edge so its a gentle reduction in lightness as it goes to the left. Then do the above.

I don't know if that dark patch to the right of the image is natural or an artefact. It was in the data before I did anything. Perhaps some artifact from previous gradient handlings. It does not look like its a natural part of the image.

Greg.

[Tony_ (Tony)]

Hello Graham,

Here's my effort with fitswork.
I think I have lost more of the blue/green nebulosity compared to Greg's method. I might give that a try too.

Regards,
Tony.

[graham.hobart (Graham stevens)]

That's great Tony-thanks a lot.
Could you email me the file? graham.hobart@bigpond.com
I will have a go at the weekend as I downloaded fitsworks last night