Pangong Lake - Dodging & Burning

[Manfred M]

I'm starting to look through my images for material that I might use in competitions later on in the year.

One of the images I am considering using is this one of Pangong Lake on the border between India and China.


With the exception of some sensor dust removal and a slight straightening of the horizon (my tripod with bubble level was off a touch), everything here is accomplished through dodging and burning. This image was done in the Lab colour space.


1. Working image - after about 45 minutes of work






2. Original image imported from ACR before dodging and burning.

[DanK]

Interesting. People often use LAB in order to separate luminosity adjustments from color adjustments. Also, people usually think of dodging and burning as methods for changing lightness. It looks to me as though you burned in the A and B channels also, which is essentially a selective color intensification, not a change in lightness. Is that right?

I've never used LAB that way, but I downloaded your original and converted it to LAB in photoshop to play around with this. I couldn't get the colors as you did, but using the B channel, I was able to increase saturation in the water and ice in a halfway reasonable way without affecting the mountains.

[pschlute]

Nice image Manfred and nice example of bringing out the best in an image.

I like the way the dark sky and water/ice funnel our view to the gap in the mountains.

[Manfred M]

Interesting. People often use LAB in order to separate luminosity adjustments from color adjustments.

That is my understanding as well. Much like dodging and burning using the Luminosity blending mode in an RGB colour space.

Also, people usually think of dodging and burning as methods for changing lightness.

How is this different than making localized adjustments in the Luminosity layer?

It looks to me as though you burned in the A and B channels also, which is essentially a selective color intensification, not a change in lightness. Is that right?

At this point I used a single global change to the a* and b* channels to increase the colour intensity. I haven't done any local adjustments there yet. Those are next on my list. I want to play around with the colours in the water and in the ice to see how that impact the overall look and feel of the image. Those are my next steps.

I've never used LAB that way, but I downloaded your original and converted it to LAB in photoshop to play around with this. I couldn't get the colors as you did, but using the B channel, I was able to increase saturation in the water and ice in a halfway reasonable way without affecting the mountains.

The clouds and mountains were my first priority in the retouch. I wanted to bring out the texture in the clouds as well as the details in the mountains (there is some in-process sharpening in the mountains and in the ice). I normally do not use in-process sharpening on clouds and water, but will be looking at the water to see if it might work for a change. It was quite windy so there was a lot of chop on the water and a bit of an edge might work well here.

[DanK]

Thanks. My point was that in LAB space, you can use the tools that people usually use for lightness--like dodging--on color as well, by choosing the A or B channel. I had never thought about that before. It's something I will have to play around with.

That is my understanding as well. Much like dodging and burning using the Luminosity blending mode in an RGB colour space.





How is this different than making localized adjustments in the Luminosity layer?





At this point I used a single global change to the a* and b* channels to increase the colour intensity. I haven't done any local adjustments there yet. Those are next on my list. I want to play around with the colours in the water and in the ice to see how that impact the overall look and feel of the image. Those are my next steps.





The clouds and mountains were my first priority in the retouch. I wanted to bring out the texture in the clouds as well as the details in the mountains (there is some in-process sharpening in the mountains and in the ice). I normally do not use in-process sharpening on clouds and water, but will be looking at the water to see if it might work for a change. It was quite windy so there was a lot of chop on the water and a bit of an edge might work well here.

[rpcrowe]

Great work Manfred! This really gives the impression of how cold it was at the moment of shooting! I also like the inclusion of the little bit of ice bound shore (?) image right bottom which gives me a boundary for the shot...

[Manfred M]

Great work Manfred! This really gives the impression of how cold it was at the moment of shooting! I also like the inclusion of the little bit of ice bound shore (?) image right bottom which gives me a boundary for the shot...

Thanks Richard:

Cold - yes it was -26°C / -15°F at the time we were shooting, very windy and at high altitude (the surface of the lake is at 4250m / 13,940 ft), Challenging shooting conditions for sure and hauling around a camera and tripod (I was shaking so much, I needed it to hold the camera steady).

Definitely a challenging shoot.

Apparently the area is absolutely packed with Indian tourists in the summer as a well known Indian motion picture, "The Three Idiots", had significant portions shot here.


https://en.wikipedia.org/wiki/3_Idiots

I think it was worth it as I got a lot of good shots at this location... I don't plan to be back any time soon....

[Manfred M]

Thanks. My point was that in LAB space, you can use the tools that people usually use for lightness--like dodging--on color as well, by choosing the A or B channel. I had never thought about that before. It's something I will have to play around with.

I've done some selecting colour enhancements using the a* and b* channels in conjunction with a layer mask. I've always made identical adjustment in both channels and set the highlight and dark colour points to the same values, otherwise one introduces a colour cast.

[DanK]

I've done some selecting colour enhancements using the a* and b* channels in conjunction with a layer mask. I've always made identical adjustment in both channels and set the highlight and dark colour points to the same values, otherwise one introduces a colour cast.

I was deliberately playing with color casts--in particular, working on the b channel to make the water and ice a bit more blue.

this is intriguing. The luminosity-only editing in LAB mode is simple enough to replicate in RGB mode, but color editing entails a very different approach. I don't know that I could replicate one in the other at this time.

[xpatUSA]

Originally Posted by Manfred M I've done some selecting colour enhancements using the a* and b* channels in conjunction with a layer mask. I've always made identical adjustment in both channels and set the highlight and dark colour points to the same values, otherwise one introduces a colour cast.

I was deliberately playing with color casts--in particular, working on the b channel to make the water and ice a bit more blue.

this is intriguing. The luminosity-only editing in LAB mode is simple enough to replicate in RGB mode, but color editing entails a very different approach. I don't know that I could replicate one in the other at this time.

Dan, I'm not sure what is meant by "replicate" in this context but these links might be of interest even if you don't use RawTherapee (RT).

http://rawpedia.rawtherapee.com/RGB_and_Lab

http://rawpedia.rawtherapee.com/Lab_Adjustments

The many curves in Chapter 8 could be of particular interest if your editor has them ...

http://rawpedia.rawtherapee.com/Lab_Adjustments#Curves

Some of them can adjust particular hues in the manner that you described i.e messing with the water and ice.

[DanK]

Ted,

Thanks. I'll study these.

What I meant is just that I can easily produce virtually identical luminance-only contrast adjustments in Photoshop regarldess of whether I am working in LAB or RGB. I posted a series of photos showing that some time ago. For that reason, I don't bother switching to LAB if all I want is luminance-only adjustments; I can more simply accomplish the same thing by staying in RGB and simply selecting "luminosity" as the blend mode.

On the other hand, I still don't know how to replicate other adjustments. For example, a standard way to increase saturation in LAB mode is to use a curve, bringing in the end points of the A and B curves. If I did that, I would have a hard time producing the same effect in RGB--particularly if I adjusted A and B different amounts. So I think your link to a discussion of curves in LAB will be helpful.

Dan

[xpatUSA]

Ted,

<>

On the other hand, I still don't know how to replicate other adjustments. For example, a standard way to increase saturation in LAB mode is to use a curve, bringing in the end points of the A and B curves. If I did that, I would have a hard time producing the same effect in RGB--particularly if I adjusted A and B different amounts. So I think your link to a discussion of curves in LAB will be helpful.

Dan

I find it useful to separate the terms "saturation" and "chromaticity" in my mind, reserving the latter for CIELAB stuff. I also keep it "simple" by thinking in polar (vector) terms, rather than Cartesian (a*,b* co-ordinates). I like the LCH model to envisage a color in CIELAB. Thus if the hue angle resulting from adjusting values of a* and b* does not change, then neither does the "color", i.e. no casting occurs. For a given hue angle, increasing the length of the vector does increase the colorfulness - similarly but identically to increasing saturation in RGB.

It should be said (apropos of an earlier post) that changing a* and b* by equal amounts does not keep the hue constant unless the initial hue angle is 45, 135, 225 or 315 degs. Picky, I know, and small amounts probably don't hurt ...

[Manfred M]

I was deliberately playing with color casts--in particular, working on the b channel to make the water and ice a bit more blue.

this is intriguing. The luminosity-only editing in LAB mode is simple enough to replicate in RGB mode, but color editing entails a very different approach. I don't know that I could replicate one in the other at this time.

Dan - the key reason I have started to work in this colour space is that I find it is faster than using one of the RGB modes and switching to the Luminosity Blending mode. My workflow means I typically run between 30 and 80 individual adjustment layers (primarily just to the L channel). It's also the widest gamut colour space, so that has its advantages and disadvantages (i.e. I can get artifacts if I push too hard, much like working on an 8-bir sRGB image.

I find the colour space is extremely effective in an image like this one, where the colour range is quite constrained. I'm playing with making small adjustments on the a* and b* channel, but have to be careful as they can be quite sensitive.

[lovelife65]

Nothing to add other than the edit works and it's a beautiful image. Love the beautiful green ice.

[Wavelength]

Wonderful edit work!!!