Rho Ophiuci Cloud Complex

[avandonk]

Gathered all the decent data I had of Rho O. It was about 30+of 4 min exposures in each of RGB and about 8 at 8 min.

It seems that if the light pollution/gradients varies over a set of subs it is problematic to use any rejection. If you do then a lot of data is thrown out with the cosmic rays and satellites etc. This is far worse when trying to meld data taken on different nights. Wide fields suffer far more due to varying gradients with target elevation.

If I use rejection all the dim dust and nebulosity disappears. PixInsight can give you images of the rejected pixels and it is obvious when real data is being rejected.

What I am doing is processing different sets of data with PixInsight to see what works. It is the integration step where it is critical to set the correct rejection criteria including none!

This image is just an average of the data weighted by exposure. 10 MB

http://d1355990.i49.quadrahosting.com.au/2014_04/RhoO_RGB.jpg

The FoV is 3.3 x 3.3 degrees.


As you can see there is a lot of dim dust far clearer than I have been able to produce in the past from the same data. If you look carefully there are quite a few asteroids and satellites as well as bright cosmic rays.

I have not used any noise reduction.

Bert

[LightningNZ]

Man that is spectacular! Loving all that dark stuff!

[SkyViking]

That's looking wonderful Bert, you have really cranked the processing up a notch lately! This image is almost a reference map of the area, very nice.

I've found the PI rejection maps very handy myself. They are essential to get the maximum S/R from the data. Sometimes it's definitely worth rejecting less. :thumbsup:

[rcheshire]

That is spectacular Bert. This area fascinates me.

Interesting observation about integration rejection parameters. The rejection maps tell it all. More time experimenting. I had always considered the defaults as safe, but it seems not so? Perhaps dithering pays off here.

[Bassnut]

Somethings wrong Bert. When I click on the link and click on the resulting "+" mouse cursor, that one zoom click shows huge noise and artifacts?. The non clicked version sure is excellent though!.

[avandonk]

Rolf and Fred I carefully looked at all the various rejection methods in PI. Linear Fit Clipping takes the variation of gradients/light pollution between subs into account when rejecting.

Here is an RGB of the rejection_high image for each of the R G and B stacks. 13 MB

http://d1355990.i49.quadrahosting.com.au/2014_04/REJHIGH.jpg

Note the many asteroid tracks RBG or RGB depending on collection order, Satellites and I would guess star profile variation and cosmic rays and even varying gradients. It seems to be doing something sensible.


Here is the image without these rejected pixels. 11MB

http://d1355990.i49.quadrahosting.com.au/2014_04/RhoO_LFC_.jpg


It is obvious the image has improved. A new set of data all at eight minutes with the now stable image train would improve the S/N even more.


Here is the Lagoon and Trifid also using LFT 18MB


http://d1355990.i49.quadrahosting.com.au/2014_04/LT_LFC_L.jpg

There is faint detail in this image I have never seen before anywhere. I just hope it is not artefact.

A screen capture showing the settings in Integrate for the red data with output images. 1MB

http://d1355990.i49.quadrahosting.com.au/2014_04/Cap_LFC.jpg


I may have finally found a way to collect high quality data from light polluted sites. PixInsight's DBE and Integrate with LFC are the two modules that seem to make this possible. I still have a lot to learn, so who knows what is possible.


Bert

[CoolhandJo]

Very deep image - lovely