Once I was teaching a class and held up a picture of the Golden Gate Bridge. I asked the class what it was. They all immediately answered, “The Golden Gate Bridge!” I had to correct them, “No, it isn’t the Golden Gate Bridge. It is a photograph of the Golden Gate Bridge.” The point was that there is a big difference between the reality of the bridge a picture of the bridge. The photo brings to mind an object, but it isn’t the object. What, exactly, is the difference between a photograph of something and the thing itself? That sounds like a silly question, but I think it is helpful to discuss the answer.

For one thing, a photo is two-dimensional. The world is three-dimensional. You can’t use depth perception to tell what is near and what is far in a photo. Because nothing is nearer or farther than anything else. It’s all on the paper in front of you. You surely can’t move your head to see what is behind that bush or stand on your tiptoes to look over the fence.

The picture is static. Most of the time the thing you’re photographing is moving. OK, so the Golden Gate Bridge doesn’t move much, but the cars and walkers on it do. So do the waves and boats underneath and the birds, airplanes and occasional clouds overhead. And so does almost everything else we photograph. A photograph captures a moment in time. I guess that’s why they call it a snapshot.

There is no perspective from peripheral vision when you’re looking at a photograph. If you didn’t get it in the frame, nobody will be able to turn their heads and see it. You can’t see the grandeur of Yosemite in a photograph of Bridal Veil falls. If you were standing there, you could turn around and see El Capitan and Yosemite Falls. But not in a photo.

The color of a photograph is always different than the color of the original scene. That’s because the sun shines with an entire spectrum of colors (remember the rainbow?) and a picture uses a combination of three colors and black to fool your eye into thinking the colors are there. Not to mention the transformations that take place in your camera and your computer. Your brain doesn’t see the colors that are there, anyhow. You don’t notice that fluorescent lights are greener than sunlight or that incandescent lights are yellower. Your brain does a little photoshopping of its own to make you see the scene the way it thinks it ought to be.

A photograph cannot capture the entire dynamic range of most of the things we photograph. You can see bright sunlight and deep shadows when you look into the Grand Canyon. Your camera will either wash out the brights or make the dim areas black unless you use special techniques to capture both in different exposures.

What is the purpose of a photograph? To capture a memory? To show someone else what we have seen? I think the photo should be true to the memory, not to the original thing itself. It should portray, not only the lines and colors, but the feeling you got when you were there and saw the thing you decided to take a picture of because you wanted to remember it. So I believe you should adjust your print to reflect your memory of the scene, not the actual scene. Especially since you can’t hope to get the thing you saw onto the paper exactly as the way it really was anyhow.

The one on the left is exposed the way my camera thought it should be. Notice the lack of detail in the shadow areas and the complete washout of the sky. The center image was taken at the same time but underexposed by 1.5 stops. It shows the blue in the sky and even a few clouds. We don’t often get clouds in Southern California, so we want to make sure they show up when we do.  The right image was overexposed by 1.5 stops. If you look carefully, you can see some the beams in the ceiling of the colinnade. Our job now is to take these three images and combine them into one image that shows the elusive clouds AND the woodwork in the ceiling of the colonnade.
One way to get detail in all the important places in this image is to load all three versions into Photoshop, choose HDR from the File:Automate menu, and merge the three 16-bit images into one 32-bit high dynamic range image. This works well and is a great feature of Photoshop. I have trouble with it, though. Since a computer monitor cannot display all 32 bits of dynamic range and my printer cannot print 32-bit images, I have to compress the image back to 16 bits. Photoshop does this for me (go to the image menu, select mode:16 Bits/Chanel) and it will allow you to have some control over the compression process, but it takes more experience than I have to get the detail I want in the places where I think detail is important.
So I use layers and masks to do the deed. As with the HDR automation, I open all three images in Photoshop. Then I copy the underexposed and overexposed images onto the normally exposed image. Make sure your focus is on the image you want to copy, select the entire image (ctrl-a on my PC) and copy it (ctrl-c). Then focus on the correctly exposed image and paste the clipboard into it (ctrl-v). The new image comes in as a new layer. Repeat this for the other image. Now you have one Photoshop file with three layers, the properly exposed background layer, an overexposed layer and an underexposed layer. Right now only the top layer is visible. Select the top layer and make a hide-all layer mask. You can either do this in the Layer menu or hold the alt key and click the mask icon at the bottom of the Layers pallette. Do this for both the imported layers.
Now you should have a properly exposed background that is completely visible because the over- and underexposed layers are masked out. Click on the overexposed layer’s mask, select the brush tool and set the color to white. I generally use a fuzzy brush with about 30% opacity and 30%-50% flow. Then paint (with the mask selected) the areas that you want to make brighter. They will start to appear as you lighten the mask that is hiding the darkest areas.
Then go to the underexposed layer’s mask and do the same for the areas that are too bright in the background. If you make a mistake, either go back in the history pallette or make black the foreground color and paint over your mistake. By varying the size and hardness of your brush, you can have absolute control over the areas you want to lighten or darken. You can also use curves or levels layers to tweak the different layers or the composite image.
I tried this with the three photos above. Here’s the result:

The sky is probably a little more dramatic than I would make it for a print. I wanted to make sure you get the idea.
Have you used this technique? Do you prefer some other technique? Leave a comment and let me know.