www.melor.com      
      Project Index      



Merging two images to increase dynamic range.
by Paul Jaruszewski

The human eye can see a much wider range from light to dark than can be captured on film or digital sensor. When we look upon a scene, we can see detail in both the shadows and highlights. A camera can too, but over a much narower range from light to dark. This can be frustrating for the photographer. On a recent trip to Yosemite, in California, while at a hotel I was looking over printed material that showed some of the sights that we were going to visit in the national park. Some were exposed with a wide range of detail in both the dark and light areas. Other photos for instance were beautiful images of a waterfall but the sky above was all white. Close inspection showed no detail in the white churn of falling water in the falls. The rest of the image looked properly exposed and demonstrated the problem that you will be able to tackle after this tutorial.

The technique of using a mask to reveal or hide a layer can be used to merge two images that have differnt focus points. For example, if you are unable to get a deep enough depth of field, take two pictures at different focus points and merge the two. It should go without saying, but I am going to say it anyway. This technique needs a tripod and preferably a remote shutter release. Use the self timer and mirror lockup in a pinch. Also, if you shoot in raw mode, you can convert one image with negative exposure compensation and positive exposure compensation. The results won't be quite as good if the dynamic range is very wide, but it can be very close. Using one raw image has the advantage of not needing a tripod to get a stable set of images.

A Tale of Two Images

This is a pair of images that I shot during that trip to Yosemite. It was a beutiful scene looking through some trees into a slow moving portion of the Merced River. (Immediately East of the swinging bridge to those who want to find the spot.) The river provided a beautiful reflection of Yosemite Falls behind the visitor's center in the park. I knew that the problem would be the lower levels of light in the reflection and the higher levels of light in the waterfall and sky. This would be difficult if not impossible to capture on one frame.

This is the frame that was exposed for the sky and other highlights. If this was the only image, it would work with just a little effort in Photoshop to bring up the darker areas. Unfortunately even bringing up the areas, while the image may be quite nice looking it will still lack the detail that was captured in the shadow areas in the next picture.

 

This is the frame that was exposed for the shadows and darker areas in the reflection, rocks and trees. If this were your only image, you could still work for the magazine that was in my hotel room - but you are better than that! As you can see the highlights in the sky and waterfall are totaly lost. This is a reminder to underexpose if you aren't sure. You can pull a lot of information out of dark areas of a digital photo. On the other hand there is nothing to see in the highlights and that area is forever lost.



The Results

This is the flattened image that was the compost of the two above. The technique is very simple and will be detailed below.

This image was further edited by cloning out the branches above and working the saturation of the layers. After reflecting ( haha ) on the image, I think the saturation was a bit too much of a good thing, but...



Step by Step Instructions

  • Load your image into Photoshop. It must not be a 16 bit image because you will be using layers.
  • The images should be exactly the same size. Select the move tool then press shift and drag one image on top of the other.
  • Select the topmost layer in the layers box, and add a mask.
  • Select the mask by clicking on it's icon if it isn't already, then change to the gradient tool. The desired gradient will be black to white or vice-versa.
  • Draw a gradient. Draw a bunch of them until you are happy with your start and endpoints of the gradient. Don't forget to press shift to constrain the direction of the gradient if desired.
  • If you need to edit the curves, etc. of the individual layers, create an adjustment layer and group it with that particular layer.
  • When you are happy with the results, you can flatten the image and continue with your edits.

 

I hope this has been a useful tutorial for you. It has been fun creating it. The concept of using masks to reveal an image below is a fundamental one in Photoshop. A variation on this technique can be used to replace skies, backgrounds, etc. Remember also that you can paint black and white on your masks using other drawing tools to reveal and hide the contents of your layers. Let me know if you enjoyed this tutorial by dropping me an email or signing the guest book in my gallery.

Paul


 

      www.melor.com      
      Project Index