saidel42 Loc: NJ

Forgetting the issue that most pictures are to satisfy the photographer and her/his friends, I have a slightly technical question having to do with the auto-contrast function in Nikon DSLR. I have a D7100. Sometimes I shoot a 3-burst with different exposures for scenic images, 1-step below, measured, 1-step above. I've been playing around with them using the auto-contrast and find that all produce similar skies (measured with a densitometer in a program called ImageJ) but the other aspects of the image are not the same. Each has a slightly different feel. I tend to go with the 1-step below because it seems to have the most edge definition (and then I measured the density across edges and it has the greatest pixel difference at edges. So what am I asking...How does auto-contrast work? and I guess psychologically, why do those scenes with the greatest contrast appear more satisfying? (I know....I could use manual contrast enhancement and I do but this issue has been bugging me.)
Needless to say, thanks for any thoughts.
Bill

Strodav Loc: Houston, Tx

I have been invested in Nikon for years and your question is interesting to me as it is a function I didn't know existed. So, I checked it out and found it under PHOTO SHOOTING MENU (camera) -> Set Picture Control. I believe it allows you to customize jpg photos to your liking. I shoot raw, which is probably the reason I didn't know it existed. So thanks for pointing it out to me

As to the 2nd part of your question, I don't know how auto-contrast works, but I can tell you that the human visual system perceives increased contrast as sharper. So, copy and image in PP and increase the contrast, especially in the mid-tones, on one of the images and look at them side by side. The one with higher contrast "looks" sharper even though the frequency content (sharpness is measured in lines / inch) is identical.

Ysarex Loc: St. Louis

Auto-contrast? Do you mean the ADL (active D Lighting) function of your camera?

Joe

Bill_de Loc: US

Picture controls work just as well with NEF files if you use Nikon software. If you shoot JPG they are baked in, but JPG images are derived from the RAW data.

The selection in Nikon software used to say "as shot", but now says 'as recorded'. Since it is not baked into the RAW file it could produce a good starting point.

---

saidel42 Loc: NJ

I wasn't too clear of course. The auto-contrast is in photoshop, not in the camera. (I've use PS 3+ for years and some steps have become automatic (not auto-contrast). I just merged them with the camera settings in my head. (I've been testing out ON 1 recently to see if it will easily replace PS3. )

I thank each of you for rsvp'ing. I actually work in the area of visual physiology. I think I am actually asking the relationship between intensity and contrast. A tough question when PS sets up an auto contrast, it does appear to react to the average I.

Strodav Loc: Houston, Tx

The way I try to understand what PS is doing to my images is to create a test image using a paint program (in my case I use corelDRAW) then measure the results.

Relationship between intensity and contrast? Do you mean how a delta L value (Lab color space) is perceived as contrast?

bleirer

I don't know the answer, but if you watch the histogram as you move the contrast slider way left and way right, you see what is happening. The more left you slide the more bunched into the center and taller the histogram gets, meaning no bright and no dark pixels, but more pixels jammed into the fewer gradations of middle tones. The more right you slide the more it stretches out both directions from from the center so the middle tones are spread apart at the expense of the darkest and brightest. I bet auto is like moving the slider until whites are as high as possible with no blown whites and blacks are as low as possible with no blown blacks.

Strodav Loc: Houston, Tx

Thanks for the clarifications. I tried it and, I believe, I am seeing what you are seeing. First, it is important to note that the horizontal axis is dark (0) to light (100) and the vertical axis is the number of pixels of a given level. So when you decrease contrast, dark pixels move right towards the center and light pixels move left towards the center, as a result the number of pixels in the center increases, hence the increase in "amplitude". When I used auto, PS decided to increase the brightness, i.e, shift the entire histogram a bit to the right, then increase contrast by spreading the histogram, i.e., moved dark / shadows to the left and light/highlights to the right. It did not spread 0 - 100, but more like 5 to 80. When the contrast increased, the "amplitude" of the histogram went down because the number of pixels at a given level got spread out as well. Hope this helps.

bleirer

I notice also that it seems to stretch the midtones more than the two ends. If you look at what the slider do (you can just click and slide right or left directly in the histogram rather than using the sliders) you can see that there is overlapping going on on in the different zones, the center overlaps with both the highlights and the shadows, so the contrast slider is working that overlap.