The human brain offers a marvelous control over scenes with differing exposure levels, unlike a camera. The viewer looks from the dark area to the light area, building a composite mental image with both areas correctly portrayed. This technique attempts to achieve a similar result using photography, where the combination of two correctly exposed images produces a composite image where all areas of the image are correctly exposed.
Neutral density filters
These are a common fix for this problem used mainly by landscape photographers. Their main disadvantage is they offer only a fixed area of control over different exposure levels and this is unsuitable for this example where the darker areas surround the bright window views.
Taking the Photographs
Take two pictures of the same scene, using a tripod so they are identical in composition. In one set the camera exposure for the interior and accept that the windows will be overexposed. This may require spot metering or some other metering method to ensure the exposure system ignores the bright window area and correctly exposes for the darker interior.
Keeping the aperture the same, for consistent depth of field, alter the shutter speed to vary the exposure. This means the manual is preferable giving the photographer total control over both of these critical camera settings. Now meter on the bright window area and take the second shot concentrating on the view through the window.
There are now two good pictures, one of the interior and one of the views looking out of the windows.
Combine in Photoshop
This tutorial uses Adobe Photoshop CS2, but similar techniques should be available in other good image editors.
- Do identical post processing from Raw or JPEG originals and create PSD, or any lossless file format, copies of both images and save them in a working directory.
- Open the image showing the correctly exposed window views and make a rectangular selection that includes the window area. It should be a full width selection starting from the top of the image to make it easier to line the selection in the next step. Copy the selection to the clipboard
- Then open up the image with the correctly exposed interior and paste the selection from the other image. Then adjust the position of the selection so it is the same position as in the other image. In CS2 this is created as a new layer as shown in the CS2 screen shot below.
- Then select the views through the windows using the magnetic lasso tool. It may be slightly slower than the magic wand tool but it allows precise positioning of the selection area. Then make the new layer active using the layer dialog box. Now create another layer using the menu commands layer-> new ->layer via copy. This copies the selected areas to a new layer 2. Turn off the layer 1 by clicking on the eye and check how the background and layer 2 combine in normal blending mode.
- Where the selected area overlaps the background layer it can be fine tuned with the eraser tool operating on the top layer. Remember this when making the selection; err on the generous side.
- Finally delete layer 1 and flatten the image to combine the top layer and the background layer. Then normally add sharpening and other normal post processing of the composite image.
- Then there it is, an image with dark areas and bright areas correctly exposed as the brain interprets the overall scene.
For more detail about selecting areas in images try this article
Join the Conversation