Orquid "Phoenix"

Orquid "Phoenix"

by ceinmart

submit your photo


Hall of Fame
View past winners from this year

Please participate in Meta
and help us grow.

Take the 2-minute tour ×
Photography Stack Exchange is a question and answer site for professional, enthusiast and amateur photographers. It's 100% free, no registration required.

I took this picture after looking at the histogram from my camera. Supposedly the details of the main object at the center was not clipped, but it is surely too dark to see it clearly. So i used photoshop's shadow/highlight command to lighten it. Is this the proper way of photographing it? or should I have increased the EC before taking the picture so that the picture of the main object would be lighter?

Original Photo enter image description here

Lightened photo by photoshop enter image description here

share|improve this question
1  
See How to choose the correct exposure? for the general answer to this question. –  mattdm Feb 26 '13 at 20:19
3  
This seriously needs color correction too. –  Olin Lathrop Feb 27 '13 at 0:00
add comment

6 Answers

I think you did what you could in the camera within reason. I looked at your top picture, and the brightest areas were nearly white. As Russell said, you can get away with a little clipping of highlights, but that would only add 10% or so. The darkest area was also completely black, so the exposure was about optimal for capturing the whole scene. Your only choice in-camera would have been to not capture the whole scene by letting some of the highlights clip and therefore getting better shadow detail.

However, you are probably OK even so. I started with your top picture and cranked up the low end in a sortof exponential slosh quite seriously. I also fixed that annoying blue cast, figuring the background of the sign was a good enough gray reference:

With this rather extreme low end brightening, we start to see quantization in the dark areas. But, I started with the 8 bit JPG you posted. You didn't say what camera you were using, but figuring 12 or 14 bit readings, you have 4-6 f-stops of additional dynamic range at the low end. Even figuring 1 or two bits lost to noise, it looks like the original has enough detail to make a presentable 8 bit picture.

share|improve this answer
    
where did you get the 8 bit? on my computer it says the bit depth is 24...? –  user133466 Feb 27 '13 at 12:50
    
That would be 8 bits per color, 8 for Red, 8 for Blue, 8 for green. –  Paul Cezanne Feb 27 '13 at 15:57
add comment

Increasing EC would result in less noise in the shadows of this image, improving the detail that was captured, at the risk of possibly losing detail in the brightest areas of the scene.

It's always a tradeoff, I think in this case you could have used EC and maintained detail in the sky for a slightly better image.

See: What is "Expose to the Right"?

The best approach if you're really concerned is to always bracket the exposure.

share|improve this answer
    
thank you matt! with this picture as it was taken, is the shadow/highlight command the proper way to process the image? what would you have done to process this picture before printing? Thanks! –  user133466 Feb 26 '13 at 16:15
    
@user133466 There many different ways, curves, levels, fill-light. Shadow/highlights is one of the more advanced methods, along with fill-light (if your version of photoshop has it). Ideally shooting RAW if your camera offers it is the best way to correct exposures, which is always going to be necessary sometimes. –  Matt Grum Feb 26 '13 at 16:20
add comment

The dynamic range of the scene exceeds the capability of your camera and some sort of compromise or adjustment is needed. But, you know that :-).

I consider that your blacks are so close to clipping that it is almost certain that you are losing information at the 'bottom end' even if the camera indicates that you are not - see image below. . Photoshop can make what you have look more acceptable but it cannot restore what you have lost - although it may look like it has done so, which may be good enough.

The segments of sky at top left are your 'brightest' areas and they do not appear to be clipping - but you have maybe only half a stop of headroom, which is not going to be enough to bring your dark areas up adequately just by shifting exposure as a whole. Relatively few people would notice if you did clip the sky highlights whereas the foreground underexposure will definitely be noticed, so if the only choice for in-camera adjustment was exposure shifting, increasing exposure by 1 stop or maybe more, would help.

You have not specified camera and model (always a good idea with this sort of question) but if you have some means of in-camera variable exposure/contrast/... management then using it or using it more aggressively here would help. (Nikon active d-lighting, Sony DRO, ...).

If you can't extend the camera dynamic range to accommodate the scene you may be able to subtly adjust the scene to fit the camera's capabilities. A whisper of fill-flash would work wonders here if flash is acceptable. This is the sort of place where, used judiciously, the much derided on-camera flash can be put to good use. It need not be obvious that you have used fill flash, while bringing the foreground dark areas up enough to render more detail and not really touching the brighter more distant areas and so effectively compressing your dynamic range.

If I was attempting to improve this 'out of camera' with my Sony A77 I'd try setting DRO ("Dynamic Range Optimiser") to maximum, expose so as to just clip the sky highlights and then see if that was good enough. If that was not good enough then accepting more sky clipping or adding modest fill flash would be next. If DRO maximum was more than good enough it could be set to auto (which will tend to minimise effect while avoiding clipping) or manually setting it to a level that worked well enough.

Using any adjustments - in camera or post processing, needs to be done with an awareness of possible undesired consequences. DRO and other manufacturers equivalents distort reality by selectively processing areas within images and this can lead to noticeable artefacts. Especially so if you know what to look for. Sony's version, which I' most familiar with, when set to maximum starts to eliminate subtle variations of amplitude in large areas that are almost constant amplitude in one or more colour channels. I've found that this can wreak havoc with the appearance of Polynesian flesh tones, especially when using flash. (Brown faces tend towards looking like coffee coloured slabs of skin).

In camera multi-shot HDR, if available, would deal with this scene with ease, but that may not be available to you.

enter image description here

share|improve this answer
add comment

Interesting shot.

It'd be helpful to know the actual camera settings, because the histogram clearly didn't tell you what you needed to know.

For example, where are the points of focus and did your camera actually meter there? What were your ISO and aperture settings? Reliance on one piece of the puzzle isn't ever enough.

My general rule (which may be backwards, actually, but it works for me) is to expose for the shadows UNLESS there is a high contrast such as here. Then, you want to expose for the middle, knowing just how far you can push Lightroom/Photoshop in either direction.

However, I would have considered the real subject of the photo to be the dark statue, not the clutter in the background. So, I would have exposed for that and made an artist decision to allow the background to be greatly overexposed. That way, the detail you want on the statue doesn't require much modification,--you'd just pull down the highlights a bit on the computer.

You can't have it all without light modification, so you just have to experiment!

share|improve this answer
    
the photo was shot in the "superior-auto" mode on a SONY NEX F3. F-stop is f/9, ISO 200, shutter speed 1/100. I don't remember where was the point of focus (can i find that out from the original image?)... –  user133466 Feb 27 '13 at 12:36
add comment

When using P, AV, SV, or full auto, EC is the best way to get a quality shot before processing the RAW. Think for your camera - if the scene is bright or backlit, + a bit, if dominantly dark, - a bit.

share|improve this answer
add comment

The flash may have helped here. I've seen pros use flash even in daylight, and that's to help fill in the shadows. Try switching to Program mode and force the flash to fire.

There's also a considerable amount of room for exposure compensation. You'll lose some detail in the highlights due to clipping, but I'd say that in this shot, the clarity of the shadows is more important than highlight detail.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.