Pages: [1]   Go Down

Author Topic: exposure adjustment vs recovery  (Read 7215 times)

button

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 427
    • http://
exposure adjustment vs recovery
« on: April 04, 2008, 12:17:17 pm »

Hello everyone,

My question applies to a RAW image that, when opened in AL or ACR, shows slight highlight clipping (not completely blown).  What is the difference between simply dropping the exposure slider to bring those highlights back (and then correcting the then darkened image with brightness or curves adjustments), as opposed to using the reovery slider to do so?

Thanks,
John
Logged

Panopeeper

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1805
exposure adjustment vs recovery
« Reply #1 on: April 04, 2008, 12:34:46 pm »

The "exposure correction" is a linear adjustment, it affects the entire range. "Recovery" reduces only the top range.
Logged
Gabor

button

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 427
    • http://
exposure adjustment vs recovery
« Reply #2 on: April 04, 2008, 12:37:18 pm »

Quote
The "exposure correction" is a linear adjustment, it affects the entire range. "Recovery" reduces only the top range.
[a href=\"index.php?act=findpost&pid=187034\"][{POST_SNAPBACK}][/a]

So, given my hypothetical problem, which approach would you use?
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
exposure adjustment vs recovery
« Reply #3 on: April 04, 2008, 12:46:28 pm »

Hi!

Highlight recovery tries to calculate color and density from the unclipped channels. Keep in mind that ACR/LR has access to all information from the sensor. Doing white balance the channels are shifted, for that reason more information may actually be available then you can see.

Erik

Quote
Hello everyone,

My question applies to a RAW image that, when opened in AL or ACR, shows slight highlight clipping (not completely blown).  What is the difference between simply dropping the exposure slider to bring those highlights back (and then correcting the then darkened image with brightness or curves adjustments), as opposed to using the reovery slider to do so?

Thanks,
John
[a href=\"index.php?act=findpost&pid=187029\"][{POST_SNAPBACK}][/a]
Logged
Erik Kaffehr
 

Panopeeper

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1805
exposure adjustment vs recovery
« Reply #4 on: April 04, 2008, 01:16:46 pm »

Quote
So, given my hypothetical problem, which approach would you use?
This depends on the actual image. Pulling down the highlights only reduces the contrast at the high end. This may or may not pose a problem, depending on the "distribution". The histogram shows, what can be done to achieve the best result; this depends on the other settings, mainly on the contrast setting.
Logged
Gabor

button

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 427
    • http://
exposure adjustment vs recovery
« Reply #5 on: April 04, 2008, 03:10:31 pm »

Quote
This depends on the actual image. Pulling down the highlights only reduces the contrast at the high end. This may or may not pose a problem, depending on the "distribution". The histogram shows, what can be done to achieve the best result; this depends on the other settings, mainly on the contrast setting.
[a href=\"index.php?act=findpost&pid=187046\"][{POST_SNAPBACK}][/a]

Hmmm... this is more complicated than I thought.  Would you mind giving me some examples, if feasible?  Up until this point, I would simply use the recovery slider to tame those slightly overexposed (not totally blown) highlights.  What I'm gathering is that the highlight recovery tool compresses highlights, in addition to reconstructing lost data based on the intact channels.  Is this correct?

Thanks,
John
Logged

button

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 427
    • http://
exposure adjustment vs recovery
« Reply #6 on: April 04, 2008, 03:27:12 pm »

I guess the heart of the matter is this:  If highlights are overexposed (but not blown), does the use of the recovery tool introduce artificial data, or does it simply compress the highlights?  If so, then using the exposure tool to bring down the overall exposure and then "fixing the historgram" with the curves tool seems like the better solution.

John
Logged

NikoJorj

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1082
    • http://nikojorj.free.fr/
exposure adjustment vs recovery
« Reply #7 on: April 04, 2008, 04:49:18 pm »

Quote
What is the difference between simply dropping the exposure slider to bring those highlights back (and then correcting the then darkened image with brightness or curves adjustments), as opposed to using the reovery slider to do so?
[a href=\"index.php?act=findpost&pid=187029\"][{POST_SNAPBACK}][/a]
The latter guarantees to touch nothing below the highlights (you may have to slightly lighten the "light tones" of the curve in some cases, but nothing more).
The former can also achieve the same goal, or very close, after more adjustments ;o).

I'd think that in both cases, ACR/LR will try to guess the missing data the same way... but we'd need an insider to be sure of that.
Logged
Nicolas from Grenoble
A small gallery

madmanchan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2115
    • Web
exposure adjustment vs recovery
« Reply #8 on: April 04, 2008, 06:15:07 pm »

Both negative exposure compensation (setting the Exposure control to a negative value) and the recovery slider will perform highlight recovery if needed. The primary difference between the tools has been pointed out earlier in the thread: EC will affect the entire range (including the midtones, where it's most noticeable) whereas recovery is limited to the top range.

Using only the recovery slider is similar (but not identical) to using negative EC but increasing Brightness. In both cases, you are effectively compressing the highlights, and the downside is that you may lose fine tonal distinctions in the highlights.

If you use only negative EC then you'll generally see more tonal distinctions in the highlights but the midtones will become darker and you'll see less tonal distinctions in the shadows. There's a tradeoff here: whenever you stretch tonal values in one place you'll compress tones elsewhere (at least when applying a simple 1D tone curve).
Logged
Eric Chan

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
exposure adjustment vs recovery
« Reply #9 on: April 05, 2008, 02:33:56 am »

Hi,

My practical approach:

1) Try to get it right in exposure (check histogram)
2) Check histogram in Lightroom
a) Reduce exposure if I can without losing shadows
 Use recovery to improve separation in highlight
3) Adjust blacks
4) Add fill light
5) Adjust brightness
6) Adjust contrast
7) Adjust vibrance

After this:

1) Reconsider image
2) Possibly iterate 2-7, preferably after taking a snapshot
3) Doing adjustment with TAT for tone curve and saturation

A hint is that you can always try "auto", it often gives decent setting for exposure and recovery. If you use auto you would probably need to adjust brightness and add fill light if needed.

Best regards
Erik

Quote
Hello everyone,

My question applies to a RAW image that, when opened in AL or ACR, shows slight highlight clipping (not completely blown).  What is the difference between simply dropping the exposure slider to bring those highlights back (and then correcting the then darkened image with brightness or curves adjustments), as opposed to using the reovery slider to do so?

Thanks,
John
[a href=\"index.php?act=findpost&pid=187029\"][{POST_SNAPBACK}][/a]
Logged
Erik Kaffehr
 

ajtaylor

  • Newbie
  • *
  • Offline Offline
  • Posts: 47
    • http://
exposure adjustment vs recovery
« Reply #10 on: April 05, 2008, 03:04:53 am »

Quote
I guess the heart of the matter is this:  If highlights are overexposed (but not blown), does the use of the recovery tool introduce artificial data, or does it simply compress the highlights?  If so, then using the exposure tool to bring down the overall exposure and then "fixing the historgram" with the curves tool seems like the better solution.

John
[a href=\"index.php?act=findpost&pid=187081\"][{POST_SNAPBACK}][/a]

Bruce Fraser's book on ACR asserted that ACR attempts to magic up any missing data, based on the channels that it has data for. i.e. if the red channel is blown, but G&B still having workable data, it will try to best guess data for the red channel. Obviously the degree of accuracy decreases if you have 2 blown channels, and if you've got lost all 3, then it's just a work of fiction.

I don't know if that is still the case - this was for ACR 2.x I believe. This was, I think, talking about using the exposure slider to recover highlights. I can't remember if ACR 2 had a recovery slider or not.
Logged

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
exposure adjustment vs recovery
« Reply #11 on: April 05, 2008, 09:06:05 am »

Quote
The "exposure correction" is a linear adjustment, it affects the entire range. "Recovery" reduces only the top range.
[{POST_SNAPBACK}][/a]

These [a href=\"http://luminous-landscape.com/forum/index.php?showtopic=24354&view=findpost&p=186076]Tone Curves[/url] from a post in another thread show the difference between the use of the exposure and recovery sliders to recover highlights. As is apparent, the tone curves are quite different in the midtones and shadows.

Bill
Logged

button

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 427
    • http://
exposure adjustment vs recovery
« Reply #12 on: April 08, 2008, 03:59:17 pm »

Thanks, everyone, for the comments.

John
Logged

rainer_v

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1194
    • http://www.tangential.de
exposure adjustment vs recovery
« Reply #13 on: April 15, 2008, 03:33:37 am »

----
« Last Edit: April 15, 2008, 03:33:58 am by rainer_v »
Logged
rainer viertlböck
architecture photograp
Pages: [1]   Go Up