2170, the picture...

[Pompey Monkey]

1 minute ago, ollypenrice said:

Thanks Paul. It's the colours which make this target, hence the insane amount of data collection.

I never answered Paddy's question and the answer is 'both.' I did run LHE to perk up the background contrasts and in the second version tried Background Enhance. I don't seem to have the LargeScaleStructureEnhance Barry mentions. In Ps I used Local Contrast Enhance from Noel and my own custom kinks in curves. (I hope it's OK to talk about  our kinks and curves on SGL...)

We'll see what Tom makes of this data. I shall probably be humiliated!

Olly

I don't have the " LargeScaleStructureEnhance " either. There is a DarkStructureEnhance script there though, I wonder if that's the one that @Barry-Wilsonactually meant?

[ollypenrice]

4 minutes ago, Pompey Monkey said:

I don't have the " LargeScaleStructureEnhance " either. There is a DarkStructureEnhance script there though, I wonder if that's the one that @Barry-Wilsonactually meant?

Ah yes, I did use that on the very dark dust lanes but it doesn't do much for less contrasty dark regions.

Olly

[Tom OD]

18 minutes ago, ollypenrice said:

Thanks Paul. It's the colours which make this target, hence the insane amount of data collection.

I never answered Paddy's question and the answer is 'both.' I did run LHE to perk up the background contrasts and in the second version tried Background Enhance. I don't seem to have the LargeScaleStructureEnhance Barry mentions. In Ps I used Local Contrast Enhance from Noel and my own custom kinks in curves. (I hope it's OK to talk about  our kinks and curves on SGL...)

We'll see what Tom makes of this data. I shall probably be humiliated!

Olly

I m reading these trying to pick up hints on how to do it

Give me a day or two, before I dare post my version.

[mftoet]

1 hour ago, ollypenrice said:

 

Thanks for the feedback and suggestions, folks. OK, time to lighten up for the New Year!! Any better?

 

Way better than your previous version! Now I can distinguish subtle dusty patches and shades in the background. 

[Barry-Wilson]

1 hour ago, Pompey Monkey said:

I don't have the " LargeScaleStructureEnhance " either. There is a DarkStructureEnhance script there though, I wonder if that's the one that @Barry-Wilsonactually meant?

There are optional processes and scripts created by enthusiasts that you can download and add to the repository, including LargeScaleStructureEnhance.  DarkStructureEnhance is something different and BackgroundEnhance produces an effect very similarly to LargeScaleStructureEnhance.  I prefer the latter process.

Now whether the success of PI with these scripts is done to 32bit processing and storage I have no idea.

[Barry-Wilson]

16 minutes ago, mftoet said:

Way better than your previous version! Now I can distinguish subtle dusty patches and shades in the background. 

 

1 hour ago, ollypenrice said:

Ah yes, I did use that on the very dark dust lanes but it doesn't do much for less contrasty dark regions.

Olly

Much better second version Olly showing the faint dust in the area.  Super.

[ollypenrice]

31 minutes ago, Barry-Wilson said:

There are optional processes and scripts created by enthusiasts that you can download and add to the repository, including LargeScaleStructureEnhance.  DarkStructureEnhance is something different and BackgroundEnhance produces an effect very similarly to LargeScaleStructureEnhance.  I prefer the latter process.

Now whether the success of PI with these scripts is done to 32bit processing and storage I have no idea.

Thanks Barry, I've found it on the net. I also have an explanation in Rob Gendler's Lessons from the Masters, it says. I'd forgotten that.

Olly

[PatrickGilliland]

2 hours ago, ollypenrice said:

PI works at 32 bit precision

I suspect there is more to it than that.  

Not sure what you cal/stack data in but I use PI for this which seems, when learnt properly to produce the best results.

Likewise multiple stretching methods that provide various optimised stretching routines it reveals more data with less effort than options I have tried in other software.

STF (left - unprocessed master lum frame)) level stretching will help find more data than auto-histogram (right) and indicate the fine details available.  Auto-histogram provides more control - knowing what's lurking within from the STF advises what can be worked on with more control after the AH route.

Or to put another way I think it is the culmination of tools designed for the job that result in more data being revealed.

I caveat that though with the fact I have not used all of the tools out there and only know those I have used...

Paddy

 

[wimvb]

The histogram in PI may look "spiked" when processing 16bit tiffs, but it never does when processing pi's own xisf files. This is due to the xisf being floating point files, as opposed to integer tiffs. The tiffs may in worst case, result in posterisation. But I think the information is present in tiffs also. PI has various tools for enhancing faint detail. The most common I use are maskedstretch and exponential transformation. The latter I learned from Alehandro Tombolini in this post

http://pixinsight.com/forum/index.php?topic=9649.msg61324#msg61324

[johnrt]

Very nice Olly, not an object I've seen before. I think there's room for a little tweak of the histogram to balance the colours up, a notch down on blue and up on red. I've checked down the back of my sofa and can confirm your dust does look very "dusty"

[Chris]

Great work Olly, never knew this one existed. Both versions look good in their own way. I'm no expert, I just like all the pretty colours  

[gorann]

I obviously have to agree with everyone that this is a masterpiece, and an unusual one, especially the second version with just the right amount of dust for my taste (today). Clearly too much dust can be distracting. Maybe someone could invent an image-format with a dust-slider so the viewer can chose the desired amount of dust........ (or post two extreme versions so the viewer can use the layer-fill slider in PS for her/his personal taste).