By David Saffir

In earlier posts, I’ve commented about the progress we’ve enjoyed with regard to new tools for photography. Here, I’ll back up those statements by briefly describing an relatively easy system that I’ve developed to make highly detailed panorama photographs. (The only caveat: Everything old is new again. About 75 to 85% of the success with this technique comes from getting it right in the camera.)

In the past, most panorama-style photographs were either cropped from a larger frame, such as a piece of 4x5 film, or photographed using a specialized panorama camera, such as the Hasselblad X-Pan.

In recent years, I started experimenting with using a digital camera to shoot multiple frames that could then be merged into a panorama. Until the middle of last year, I wasn’t satisfied with the results. I had issues with variations in brightness in the scene, lens distortion and other problems.

However, with the release of some new hardware and software tools, shooting and merging frames has become impressively consistent and requires very little extra work in post-production.

Here is a brief rundown of the techniques I use. (Look for a more detailed version to be posted on my website within the next few weeks.)

Use a normal focal length lens, or slight telephoto lens that has good flat field characteristics and little or no change in brightness in the corners.

Use a tripod that has a reasonably good camera mount that can be rotated easily. Level the tripod before you mount the camera on it.

Use a slide mount to attach the camera to the ball head. You will use this to move the center of the camera back and away from the center point of the tripod so that the camera rotates around the optical center of the lens. This eliminates parallax. (The mount shown here is made by Really Right Stuff.)

To know how far back to move the camera on the slide, look up the lens optical center, or nodal point, on the manufacturer’s website. (This info is usually found in the technical specifications for the lens.) Move the slide back from the center of the tripod an amount equal to the number provided for the nodal point. (see photograph)

Mount the camera in portrait, not landscape orientation. You’ll get great top-to-bottom coverage and when the frames are combined the result is very impressive.

Use a bubble level to make a final check of the camera and tripod by slowly rotating the camera through the range of the scene you want to photograph.

(Right now I’ll bet many of you are saying “that’s going to take too long!” but with a little practice setup takes about five minutes.)

Use manual focus, and manual exposure. Try to pick an exposure that is close to the middle of the range of exposure measurements that you get from one side of the scene to the other. Set the camera up in Aperture priority, at a moderate f/stop (f/8 or f/11 is a good starting point).

If you are shooting RAW image captures (recommended) set your white balance to manual, or to one of the lighting presets, such as daylight, in the camera. Using “auto” setting will create headaches for you later in image processing.

Shoot from left to right and expose the frames you’ll need to capture the scene. Overlap the frames by 25-35%. Be sure to use RAW capture, and do not use a polarizing filter.

That’s it for capture – the next step is ridiculously simple.

In Photoshop CS3 use the new Photomerge utility (File>Automate>Photomerge) to merge the images. Set it on “Auto”, and click OK.

Photoshop will create a merged version of your group of images. It will look a bit like a bowtie, and it will have as many layers as you have frames. Flatten the image, crop it to the rectangular aspect ratio you want, and make whatever other adjustments you wish.

To summarize: get everything level from the ground up, use manual focus and exposure, shoot RAW, and overlap the frames. Load them into Photoshop’s Photomerge, and voila! The results are amazing!

After I shared this technique with photographer Ted Dayton, he sent me back this note:

“I processed nine RAW files into 12-MP TIFFs and merged them in Photoshop to create the finished image. About 30 minutes and 450 MB later, the results were amazing!! Anybody want to buy a 6 x 17 camera?”