Welcome to the forum!

As an adjunct to the Tangents blog, the intention with this forum is to answer any questions, and allow a diverse discussion of topics related photography. With that, see it as an open invitation to just climb in and start threads and to respond to any threads.

Over exposed skin with overcast sky

rtcaryrtcary Member
edited December 2010 in general photography
With a overcast sky, I often have parts of the skin produce over exposed areas. Here are some test pictures I took with my new D700 of my grand-daughter: http://picasaweb.google.com/lh/photo/eChNLB7VeWGMHCsCq-9YaQ?feat=directlink
I did some burning in with Photoshop, however the data is not there.
1/250 @ f/2.8 ISO: 250; 28-70 f/2.8 @70 mm
It was suggested that I should use a scrim or flag. Suggestions welcomed.


  • Hey @rtcary did you shoot in RAW? If so, you should be able to recover some data using the 'Recovery' slider. If you're shooting in jpeg, then I'd suggest spot metering your grand-daughters face and meter off that.

  • Yup! RAW it was and I did pull up some of the data with "Recovery". Also, the histogram looked near perfect and my overexposure indicator showed that I had NOT strayed. A local pro likes to have something blocking any overhead light; likes to use doorways (I do not have a portable doorway :-) ). Just wondering if others run into this problem *or* am I making some mistake.
  • Looked at the image and to be honest I think you've nailed it. I can't see any blown highlights unless this is the tweaked version. With her face tilted like that to the sky I think that looks OK and the skin tones look great.
    If ACR and PS show no histogram blow outs then I don't think you've a problem - unless you can show us the raw and the raw tells different.
  • @robby
    Appreciate your generous comments. And granted, I am being critical - maybe overly - in the ongoing effort to improve my images. After looking at my test images again (before Photoshop), areas of the skin (e.g. bridge of the nose, forehead) are lighter so the depth of color is not what the eye sees. "Over exposed" was not the best term to have used; instead, I would like to have slightly more uniformity in the depth of color by learning how to control the lighting.
  • @rtcary I agree with what @robby said, I think the image looks great with the nice soft light. I have come across the problem of having parts of the face overexposed (captured in RAW) and then when I pulled the recovery slider up in ACR found that the blown-out highlights had this wierd white/washed out looks (I assume from there being no data captured in that tonal range).

    However, in your instance, I think it looks completely natural :)
  • I think the exposure is great, what your seeing is uneven overhead lighting, a large difuser panel would certainly even it out and give you a softer light.
  • Neil vNNeil vN Administrator
    Todd, you mention that you're getting over-exposed areas on her skin.
    The image looks fine to me.

    What I think you're seeing is perhaps the 'over-exposure' warning in ACR or Lightroom?
    It could be that one of the channels is blowing out slightly.
    For me, this isn't really a problem, as long as I like the overall exposure and skin tones.

    What could be contributing here, is if you are using the sRGB color space.
    It has less leeway than Adobe RGB in how it renders colors. So you could very well find you're NOT getting the over-exposure warning if you select Adobe RGB.

    Either way, the image there looks fine to me in terms of exposure.
  • Actually MattH described it. Sorry I was not clearer. In my next test, I'll try using my diffuser. Thank you all.
  • DanDan Member
    edited January 2011
    This is something I have been working on and will be adding to my new blog here soon I’m sure. I like to keep my skin tones very consistent so I have come up with a way using the zone theroy. Basically what I do is convert the photo to B&W and use the Eyedropper Tool to check my tones compared to the zone I want the skin to be placed in. As you can see in this screen grab ;


    Now to get just the tones in the face you can crop it like this;


    You can look at the histogram and push up the recovery, drop back the exposure, add fill whatever it takes to bring the facial tones down to the zone you want. In Lightroom there are three vertical lines dividing the Histogram, each line represents 2.5 zones (respectively). So, I just added a little recovery and pulled back some on the exposure until the graph dropped off at zone 7.5. Now you have to be careful when pushing a photo a lot, Recovery can add artifacts you don’t want and waste a lot of time trying to clone them out. If I was on a regular shoot, I would shoot a test shot just like this crop and check my Histogram for proper tonal placement. Sometimes it helps to hold a big black flag behind the subjects face to help you read the proper tones on the Histogram.


    And here you have better skin saturation and proper tonal placement. I say proper only if I knew where her skin tone should be placed. I just started a blog at http://danielgreenphotography.wordpress.com/ and I would love it if anyone with questions about this came over to read and share. I don’t have this posted yet, since I just started my blog last night.

    Please come join me at [Daniel Green Photography
    Promote Your Page Too]

    Oh and by the way; Niel I LOVE YOU MAN! I would be happy to just learn what you may have forgotten. I know I have said that to you before, but I mean it!
  • Hi,
    What metering mode you used?
    The matrix metering mode is having some kind of protection to skin tones.
  • If that question was directed to me, I used centered weighted on a D700.

    ISO: 250
    1/250 @ 2.8

    70 mm with Nikon 28-70
  • @ rtcary: sorry to adress you with my first post.
    I think, your metering was not 100% correct and that led to this error.
    Kindly do the same exposure with same time with Matrix mode. I think, you will over come form this error.

    Please note that the said bridge of the nose and forehead are a bit closer to sensor plane and it will be almost away from the evaluative area of centered weighted circle that used by Nikon Centered weighted metering bias.

    Please look at this link and more references.
Sign In or Register to comment.