My Manual HDR stack of the eclipse - Plus New Processing Method

[DavidTrap]

Here is my offering to the many eclipse photos being posted.

I'm not happy with my focus, but will use that as an excuse to go again ;)

This is merge of 4 exposures, each 1 stop apart, manually cut and merged using layers in PS. Tried the automatic HDR in PS5, but it didn't do it for me.

Also included is a Diamond ring shot.

Comments welcome, especially on how to manage the "banding".

ta
DT

Edit: Have added a slightly brighter, more contrasty stack.

Edit 2: A rough application of a method I found on the web tonight - might have to set aside some time to see what I can extract from my data.

[allan gould]

Very nice images, David - very impressed and it looks like you had a good time.

[Paul Haese]

Diamond ring is quite nice.

[Larryp]

Nice images, David!

[jjjnettie]

:) Stunning

[naskies]

Great job - looks amazing! Were the proms visible to the naked eye at all?

[gregbradley]

All 3 are very good. There is some aliasing at this resolution so I assume the originals didn't show that?

Colour banding is removable to a large degree although it seems like a rainbow effect with the high thin cloud.

I played around with it and seemed to be able to boost detail and remove the banding to a large degree.

Lovely shots. An unusual brown band on the top that is like a blur - is that a bit of motion blur rather than focus?
I saw something similar with some sun test shots I did.

Greg

[RickS]

Nice work, David!

We finished our two day drive home this afternoon. I hope to have time to break out the processing tools tomorrow.

Cheers,
Rick.

[DavidTrap]

Thanks for the comments,

Rick - glad you're home safely. Look forward to your processed images, and when you work out what needs to be done to your's, I'll seek your assistance with mine!

Dave - proms were definitely visible naked eye and looked fabulous through binoculars.

Greg - the banding I was referring to was due to the layering in PS. I tried to use a large feather, but more sophisticated methods might be required. The brown band is due to movement of the moon during the sequence. I was following the advice from a workshop that one should align on the corona, rather than the moon. I reduced the opacity of the layers in an attempt to reduce the banding, but this artefact was the consequence. What methods did you use to improve "detail"?

DT

[gregbradley]

Greg - the banding I was referring to was due to the layering in PS. I tried to use a large feather, but more sophisticated methods might be required. The brown band is due to movement of the moon during the sequence. I was following the advice from a workshop that one should align on the corona, rather than the moon. I reduced the opacity of the layers in an attempt to reduce the banding, but this artefact was the consequence. What methods did you use to improve "detail"?

DT[/QUOTE]

If you push the image the banding is circular and in all outer areas past the moon. That's why I figured its from high thin cloud and is a rainbow effect.

The banding can be reduced in several ways. Lasso the moon and first little bit of the corona and feather and inverse select. Then switch to lab mode then gaussian blur a and b channels.

Another way (didn't try this but it probably would work) would be to desaturate the outer parts of the image as above. The corona appears to be only white so this would not damage the image.

Another way is gradient handling like Gradient Xterminator or other more advanced Photoshop gradient handlings. But the lab method worked best.

The gradient handling also increase detail in the corona and reduces the bright halo which is just a glow and in parts a bit overexposed. You see a bit more detail.

Greg.

[DavidTrap]

Added a new image with a different processing method that extracts more detail from the corona.

DT

[Phil Hart]

Nice images David. There's no easy way to do this.

You might like to try this software Geoff Sims posted on my thread

http://www.gva-hamburg.de/sofi2006/fitswork/sofi_fitswork_uk.htm

Haven't tried yet (still in the Daintree) but it looks promising!

Phil

[RickS]

That's looking even better, David!

Thanks for the pointer to the software, Phil. I had a quick play with PixInsight today but haven't found a way to automate the registration of the images yet. Hopefully, Fitswork will do the job.

[gregbradley]

That seems like a nice improvement David. The 2nd one seems to have a dust donut on the lower left? That might healing tool out.

You've got some of the nicer shots here.

Greg.

[geoffsims]

David,

Your latest image is looking great! I forget the best way to do it, but one thing you may wish to try now is changing the colour of the background sky (currently black) to a deep blue. This typically gives a much more pleasing image, and is more true to the actual colours you see visually.

Geoff

[DavidTrap]

Thanks again for the comments Gents,

Phil - I found the recipe at the bottom of that link. Haven't tried the software on the original page.

Geoff - I like that suggestion.

Rick - the registration relies on finding a star - mine are slightly out of focus, so no stars...

DT

[multiweb]

Superb shots David. Best corona so far. :thumbsup::thumbsup::thumbsup:

[Paul Haese]

Nice David, I like the last one best.

[iceman]

Great! The last one is excellent. How did you do that?

[DavidTrap]

Thanks again,

Greg - there are about half a dozen dust spots that I've found in Lightroom - a batch process will hopefully sort them out.

Mike - I used one of the links at the bottom of the website that Phil suggested. I don't know how or why it works, but it does improve the detail. :shrug:

DT

[RobF]

Great stuff David. Glad to see you bagged some decent images :)

[DavidTrap]

Thanks Rob,

Can't wait to try the Fitswork software on the whole stack of images. This technique seems to deliver the goods.

DT

[ourkind]

Very nice images David. Thanks for sharing!

[firstlight]

Nice work David. We missed the second diamond ring, thanks for sharing.