Hanging out at Sundown

[Stagecoach]

I was experimenting last night with bracketing during the post sunset period and although the sky and light was pretty drab I have a number of sets to use in exposure merging.

This was one of the trials, three images HDR merged in ACR, not too much difference than I could achieve from one exposure but quicker.

Did I catch a hanging at sea

No 1



Comments and suggestions welcome as always.

Grahame

[DanK]

Interesting shot.

I wonder about lightening the boat and increasing contrast in the mountains. I didn't do a careful job with selections, but this shows the general direction I was thinking about:

[Geoff F]

I prefer the second image.

After experimenting with a number of auto HDR programmes I now use manual merging using layers and masks/selections. Sometimes with various forms of auto selection (plus a little manual adaption) but mostly by brushing to create partial selection of masks.

[Chataignier]

Nice image, but i agree with Dan that the boat should be lighter.

[Manfred M]

I agree with the lighten up the boat sentiment, but also would like to the the bright sky toned down just a touch. I haven't done a particularly good job on it, but one gets the idea.

[Stagecoach]

Dan, Geoff, David and Manfred

Rather than duplicate things I'll respond to all the comments so far here.

The last thing I did to my posted version was to tone down the boat as I considered this truer to the scene as I saw it. One area I'm always aware of is over brightening (used loosely) a subject with relation to its background which in this case was only backlit. It's seems a fine balance in producing a low light after sunset scene and not turning it into one that looks more like daylight. Then of course we add personal taste into this

As for the sky, there are a number of possibilities here and even darker may work.

With respect to 'HDR' for this image it was purely a trial using the 5 different exposures I had captured to use with manual masking if needed. But, in this example the contrast of the scene was all captured within one exposure which could easily be manipulated to get an equivalent result.

What I have since been doing with this image is experimenting with 'Luminosity Masks' to concentrate adjustments just for learning if they have a worthwhile value with this type of scene.

As for the image, I'll score it 3 out of 10 for the simple reason it's mundane and there was no decent light or colour

[Wavelength]

NIce image; lovely clouds and their nuances;i would prefer to remove the distant tug and lifebuoy to suit my taste

[DanK]

Can you post one of the original images, ideally the one that is brightest without blowing the sky? Did the image exceed the dynamic range of the camera?

For the most part, I don't understand the value of HDR when images don't exceed the dynamic range of the camera. The only time I would be tempted to use it with an image that fits within the dynamic range of the camera is if I had large areas in deep shadow and wanted an exposure with a higher ratio of signal to noise in that area. In essence, it's automating tonality adjustments using an algorithm that has nothing to do with the particular image. I find that even when I do need HDR or exposure fusion, it often gets tonality wrong, and I have to adjust the composite manually anyway.

[xpatUSA]

The last thing I did to my posted version was to tone down the boat as I considered this truer to the scene as I saw it. One area I'm always aware of is over brightening (used loosely) a subject with relation to its background which in this case was only backlit. It seems a fine balance [between] producing a low light after sunset scene and not turning it into one that looks more like daylight. Then of course we add personal taste into this

As for the sky, there are a number of possibilities here and even darker may work.

Grahame, it may be of interest that my fusion app allows the application of a "quality"weighting curve depending on pixel brightness, please read from here onwards for couple of sections. It does much more that just a tone curve. Hopefully you'll realize the implications for your particular images in this thread.

Maybe your fusion app has similar?

[Stagecoach]

Can you post one of the original images, ideally the one that is brightest without blowing the sky? Did the image exceed the dynamic range of the camera?

Yes Dan. This is an exposure from the sequence where the settings captured the entire contrast range without clipping. Camera Neutral profile and only a minor tweak to WB.

As mentioned in post 1 I could produce basically identical final results using just this one exposure.

No 2

For the most part, I don't understand the value of HDR when images don't exceed the dynamic range of the camera. The only time I would be tempted to use it with an image that fits within the dynamic range of the camera is if I had large areas in deep shadow and wanted an exposure with a higher ratio of signal to noise in that area.

I would agree.

In essence, it's automating tonality adjustments using an algorithm that has nothing to do with the particular image. I find that even when I do need HDR or exposure fusion, it often gets tonality wrong, and I have to adjust the composite manually anyway.

I have not delved into HDR yet but am going to do some further trials with greater contrast sunset scenes to see if it gives any advantage to how I presently work them.

General comments from this shoot/trial and looking at the 12 sequences of 5 shot brackets produced ...................

Primarily my interest at present is tackling after sunset shots where you have the problem of foreground or subjects being very much in shadow. There are various ways of tackling this and I think a 5 shot bracket will now be standard practice for me however I use the exposures.

One of the problems encountered is the fast changing scene contrast during this short period and images shot just a few minutes before or after the one posted here look significantly different in a single exposure with no to minimum clipping. Another problem encountered is movement with the increased speeds.

[xpatUSA]

As mentioned in post 1 I could produce basically identical final results using just this one exposure.

No 2

The min in that image No. 2 is 4/255, the max 232/255. Working those values backwards through DPR's DR test graph for the D800 tells me the scene DR was about 7.5 EV, easily within the camera's capacity, I imagine.

[Stagecoach]

The min in that image No. 2 is 4/255, the max 232/255. Working those values backwards through DPR's DR test graph for the D800 tells me the scene DR was about 7.5 EV, easily within the camera's capacity, I imagine.

Yes image No 2 was easily captured within the DR of the camera and interesting to know that you have calculated a DR of around 7.5 EV.

For interest here is the scene taken 7 minutes before No 2. Again this one uses the same parameters as No 2 and is the one of the bracket sequence that is closest to clipping the highlights. I assume your calcs will show a higher DR in this one.

No 3

[Stagecoach]

Grahame, it may be of interest that my fusion app allows the application of a "quality"weighting curve depending on pixel brightness, please read from here onwards for couple of sections. It does much more that just a tone curve. Hopefully you'll realize the implications for your particular images in this thread.

Maybe your fusion app has similar?

I had a read of the Link Ted and can see where adjustable result weighting based on pixel exposure, contrast and saturation could be advantageous in certain situations but it's 'global'. I'm not sure if it would be useful in a scene such as this where the tonalities of subject and background are similar and what we really want to end up with is subject isolation e.g. "brighten the boat".

The only two HDR procedures I have played with recently are the ones in PS and neither have control other than on the finished merge. I'm waiting for the weather to change so I can go out and get some good test images to trial them.

[xpatUSA]

Yes image No 2 was easily captured within the DR of the camera and interesting to know that you have calculated a DR of around 7.5 EV.

For interest here is the scene taken 7 minutes before No 2. Again this one uses the same parameters as No 2 and is the one of the bracket sequence that is closest to clipping the highlights. I assume your calcs will show a higher DR in this one.

No 3

Something wrong with my method, Grahame. It is affected by exposure and therefore how the scene luminance is rendered in the posted image. My method gave a DR of 6EV for this scene which can't be right. Sorry.

[xpatUSA]

Something wrong with my method, Grahame. It is affected by exposure and therefore how the scene luminance is rendered in the posted image. My method gave a DR of 6EV which can't be right. Sorry.

I'm thinking that my simple method only works if the exposure is specifically for the good old 18% average scene reflectance and the image is rendered per dumb meter average illuminance and if the converter/editor use linear tonal rendering i.e. no curves and if the scene is within the limits of the camera DR ... glurk!

It's enough to make an old man grumpy ....

... Still thinking about it ...