Life is full of hurdles. Perhaps no more daunting are those that we impose on ourselves. Whether you pursue photography professionally or as a hobby, maintaining that creative spark is difficult. Usually we fall back on our tried-and-true approaches to photographing which can stifle our passion and lead to stale images. What prevents us from exploring the many new techniques that digital photography offers? Mostly it is fear: fear that we won’t understand something; or fear that we won’t be good at it. Yet, when we do find the courage to break out of our confining boxes, we look back at the experience thinking that it was not as hard as we originally thought and question why we did not venture out sooner. Well, cast aside your self-doubts as you find out how easy it is to produce high quality panoramas from your digital camera. There is nothing to lose and everything to gain.
Advantages
What is a panorama, how is it created, and what are its advantages? Essentially a panorama is an elongated, unobstructed view of a landscape that can emulate or exceed the field of view of the human eye (about 160 degrees). This can be a vertical image, although most panoramas are laid out in the horizontal, or landscape, mode. In technical terms, it is creating an image that has a higher aspect ratio (i.e., width-to-height) than your camera sensor. Generally, to reproduce the field of view of a human eye requires an aspect ratio of 2:1 or greater. For a full-framed DSLR, or for those people shooting 35 mm film, this means exceeding a 3:2 aspect ratio. For example, 4:2, 5:2, 6:2, and so on, all capture a wider field of view than a single frame and would be considered panoramic. If you are shooting with a point-and-shoot camera that has an aspect ratio of 4:3, then 6:3, 7:3, 8:3, etc., are panoramic relative to the single frame.
In the days of film cameras (isn’t that beginning to sound quaint?), creating panoramas required a special camera and lens to record an image on film, like a Linhoff, Horseman, Noblex, or Fuji. Lacking that, photographers would cut and paste together several single-frame prints to create a panorama. Consequently, it was only as good as the skills you learned in kindergarten. As you would expect, today the process of cutting and pasting several images is done using software and is called photo stitching. Two techniques are used. The first is by stitching together a single row (or column) of images, called single-row (or single-column). The second is by shooting/stitching multiple-row horizontal images (or multiple-column vertical images), called omni-directional. As you can imagine, the omni-directional technique is more complicated and time consuming, and for this article we will restrict the discussion to single-row horizontals.
Photo stitching can be done on aspect ratios other than panoramic, but regardless of the number of photos used – single- or multi-row layout, or the aspect ratio of the final image – they all share one thing in common. That is, they significantly increase the amount of information that is captured of the scene. In other words, they effectively increase the resolution of your camera, something most of us desire without the investment of a new camera. Let’s look at a simple example. Assume we are using a full-frame DSLR. To create a single-row panorama the camera is rotated to the vertical, or portrait, position. Now the aspect ratio is 2:3. If three images are stitched together without any loss of pixels, then the final image has an aspect ratio of 2:1 (3 x 2:3 = 6:3 = 2:1) and, in theory, there are three times the pixels. In practice, however, some pixels are discarded and you end up with something less, depending on the overlap. For the sake of argument, let’s say we end up with twice the pixel count. Then, for a DSLR with a 21 MPix sensor, that means capturing an image with about 40 MPix, or the equivalent of a $20,500 Hasselblad H5D-40 (40-MPix) medium format camera!
If you think that this is just an academic exercise, then the “proof is in the pudding,” and for photographers our pudding is the print. When I saw my first inkjet panoramic print created from the aforementioned 21 MPix DSLR, the same “Wow” factor was there as when I saw my first print from a medium format film camera and compared it to a 35 mm film camera print. Others agree, that for all intents and purposes, a panoramic print made from a DSLR camera is indistinguishable from a print made with a single-frame, high-resolution medium format digital camera. Some even claim that it rivals 4×5 film1. I’ll let you be the judge. For the first time in years, I have found a technique that captures the grandeur of the Alaskan landscape, and for me, panoramas take my images to the next level and that is why I find this so exciting.
Equipment
Before taking the individual panorama images you need to have the proper equipment. Starting with the basics, a good, stable tripod is a must. Some people have successfully made panoramas by hand-holding the camera, but you want the sharpest images possible to make your work stand out. If you are not using a tripod for most of your images, why not? Despite the advances in image stabilization or vibration reduction, a properly used tripod will always produce a sharper image than hand-holding. Also, since you want to shoot at a low ISO to minimize noise and a high f-stop to maximize depth-of-field, a shutter release (cable or wireless) should be used. If you don’t have a shutter release, then use your camera’s self-timer. Lastly, an L-bracket is needed on your camera to be able to rotate it to the vertical position. L-brackets create a more stable camera/tripod setup when shooting verticals by keeping your camera’s center of gravity over the tripod rather than flipping the ballhead’s stem over into the vertical notch. They can be purchased from equipment suppliers like Really Right Stuff and Kirk Enterprises.
Now for the specialized equipment. As we will discuss in more detail in the next section, creating panoramas requires having a level tripod so that your horizontal (or vertical) images are straight. And it requires eliminating parallax error. This means that as you create images by panning left to right, foreground objects need to stay in the same relative position to background objects. Otherwise, the finished panorama will show ghost images of foreground objects. To eliminate parallax error, the camera must rotate about what is called the nodal point2 of the lens. This necessitates being able to shift the camera body behind the center of rotation of the tripod. The best equipment I have found for doing this is the Pano Elements Package from Really Right Stuff for $360. Despite the high cost, I’m partial to RRS because not only is their equipment rugged (bomb-proof is more like it) and functional, but it is also beautifully designed. The Pano Elements Package provides a precision panning base (with a spirit level) and a nodal slide bracket, also with a spirit level. I also recommend getting the PCL-DVTL dovetail ($30) so that you can quickly attach the panning base and nodal slider to your existing ballhead clamp, rather than having to remove the clamp and attach it to the ballhead with a screw every time you want to create panoramas. The Pano Elements Package is only for making single-row or single-column panoramas. RRS’s Ultimate-Pro Omni-Pivot Package ($795) is needed for making Omni-directional panoramas.
Two other options exist for making panoramas. In Outdoor Photographer3, George Lepp discusses use of the GigaPan Epic and Epic 100 ($300-450) which are robotic camera holders for small DSLRs and point-and-shoots used to automate the process of taking dozens or hundreds of photos. The composite image contains billions of pixels hence the name GigaPan. Since that article was published, Gigapan introduced the Epic Pro ($895) for full size DSLRs. The other option is tilt-and-shift lenses, particularly the Canon 17 mm T/S and 24 mm T/S. These lenses allow you to rotate the tilt mechanism 90 degrees so that panning can be achieved in the horizontal direction. However, the range of movement is rather limited so you would still have to pan the camera body to cover a panorama.
Setup and Taking the Photos
Three steps are required for setup4: a) leveling your equipment, b) finding the nodal point, and c) adjusting camera settings. First, attach the panning base to the tripod head and mount the nodal slide to the panning base. Level the nodal slide using the spirit level. It helps if you start with a reasonably level tripod to begin with, if anything for the security of your equipment. Once the nodal slide is level, tighten down the ballhead and mount the camera in the vertical (portrait position) with the L-bracket to the clamp on the end of the nodal slide. Make sure the centering index marks on the L-bracket and nodal slide are aligned.
Second, to determine the nodal point of the lens find two objects, one in the foreground and the other in the background, and line up the objects in your viewfinder. Position the approximate center of your lens (if you are using a zoom lens then make sure the lens is zoomed to the focal length you want) over the tripod’s axis of rotation by sliding the nodal slide forwards or backwards. Now pan left. If the background object shifts left relative to the foreground element you are ahead of the nodal point. Move the nodal slide forward and repeat panning left. If the background object shifts right relative to the foreground element you are behind the nodal point. Move the nodal slide back and repeat panning left. When the nodal point has been located, the foreground and background objects will not move relative to each other when panning. Really Right Stuff has also tabulated known nodal points for specific camera/lens combinations5.
Third, choose your camera settings prior to taking the photos. Shoot raw for maximum flexibility in post-processing. Use a low ISO, preferably ISO 100, for minimum sensor noise. Decide the depth of field desired. Since most panoramas are of landscapes and a maximum depth of field is desirable, a high f-stop is appropriate. That does not mean automatically set your lens to the maximum f-stop, like f/22. Remember that the sharpest aperture of any given lens is not necessarily the highest f-stop, but one or more f-stops open from the maximum. Another way is to use the hyperfocal distance for the f-stop. Set your camera exposure system to manual. This will assure that your exposure is consistent across the entire scene. However, since panoramas can cover a wide range of light values (or EV), set your camera exposure to manual and determine the shutter speed that will keep your histogram as far to the right as possible for the brightest part of the scene. You may need to use a graduated neutral density filter if the exposure exceeds the dynamic range of your sensor. Turn off auto white balance and set white balance manually so that, like exposure, white balance will be consistent across all the photos. If you get the white balance wrong you can adjust it with your raw converter. Finally, use Live View (how did we ever do without it?) to set your focus on the appropriate part of the scene. Shut off auto focus and image stabilization (or vibration reduction). If you have mirror lockup, then activate that function.
You are now ready to shoot. Using the shutter release, take a sequence of photos which cover the scene by starting at the left side and panning right after each shot. Overlap each shot by 30-50%. Anything less than 30% and the stitching software may not be able to find enough commonality to create the composite image. The hardest part of creating a panorama is done!
Work-up
It does not matter which raw converter you use to work up the raw images, whether it is Adobe’s Camera Raw, Adobe® Photoshop® Lightroom® 4, Apple’s Aperture, or the camera makers’ raw converters. Use the work flow with which you are most comfortable. I recommend exporting your images as 16-bit TIFF, or preferably PSD, files. This is not the time to compress to an 8-bit JPEG. Even though all photos were shot in manual exposure mode, I have found that light values will range across the sequence. I have successfully changed the exposure value in the raw converter individually for each shot to match up the histograms better. Be careful that you don’t spread out the shadow areas too much revealing the underlying noise.
Believe it or not, the easiest part of creating a panorama is the stitching in Adobe® Photoshop® Elements 11 or CS6. You do not have to understand layers or masking to proceed. The following instructions are for a Mac computer. Once you have the converted images saved, from the Photoshop CS6 menu select File > Automate > Photomerge… (In Photoshop Elements 11 select Enhance, Photomerge®…, Photomerge®Panorama…) In the dialog box under Source Files, Use: select Files. Press the Browse button and locate the place/folder where your exported images reside. Select all of the image files taken for your panorama. Press Open. You should see all of your image files in the Source Files window. On the left side of the dialog box under Layout are seven choices: Auto, Perspective, Cylindrical, Spherical, Collage, Reposition, and Interactive Layout. Auto works best for most panoramas, although you might need to try Perspective, Cylindrical, or Spherical for problematic panoramas. Lastly, at the bottom of the dialog box, Blend Images Together is selected by default. If you have not corrected for vignetting in your raw converter you may want to select Vignette Removal. The Geometric Distortion Correction option is probably for panoramas of buildings and other architecture. At this point you are ready to let the computer do all of the work so press OK. Depending on your computer processor speed and useable memory, it could take several minutes for the composite panorama image to appear.
When processing is done you will have an image with the same number of layers as photos taken. Each layer will have a layer mask associated with it. If you have leveled your camera and found the nodal point, the resulting composite image will mostly fill a rectangular box. Otherwise, the individual layers will either bow out or in at the top of the frames leaving blank space. This means losing some of the image to crop it to a rectangular shape. Initially you may be disappointed with the stitching process because there appear to be jagged lines, like pieces to a puzzle, where each layer mask butts up against its neighbors. Some claim this is a flaw in the software, however, I think Adobe left it in the software to show you where each stitch is so that you can assess the quality of the compositing process. I’ve created over a dozen panoramas and have not found a matching problem yet. Once you flatten the layers (Layer > Flatten Image) the lines disappear and you will likely never be able to find the seams again. After that, crop the image to suit your goal for the panorama and finish with any additional post-processing.
Figures 1-3 provide three example panoramas. These images don’t do justice to the actual prints, however. Susitna River is a composite of five photos, Eagle River Valley ten photos, and Chugiak Winter nine photos. The prints, ranging in size from 40-61” wide x 22” high reveal detail and three-dimensionality that exceeds what can be achieved with one DSLR photo. You get the sense of being able to walk into the landscape, enhancing the ability to convey to the viewer what inspired you about the subject. Isn’t that what photography is all about – telling a story, conveying an emotion, or inspiring others to action?
Finally, there are other stitching programs available, most notably GigaPan Stitch (comes with GigaPan hardware) and Autopano. The camera manufacturers provide their own versions, and there is even a stitching app for the iPhone’s built-in camera. I can’t vouch for their ease of use or flexibility, but as you have seen with Photoshop, there is no hurdle to overcome in using it.
Tips and Hints
The following is a list of tips and hints that will make it easier for you to create flawless panoramas from the start. Some of these I’ve learned from the “school of hard knocks.”
- If your scene encompasses a wide field of view, avoid using a polarizer since its effect will be uneven across the resulting panorama. For example, if you are shooting with the sun 90 degrees to your right or left and create a panorama spanning 180 degrees, then the polarizer will be most effective in the middle of the panorama and non-existent at the left and right edges.
- Create your sequence of images fairly quickly, especially when there is movement in the scene like clouds, shadows, wildlife, people, or water. For example, I created a panorama of a tarn and was not mindful of the effect the wind was having on the surface of the water. As a result, the water was flat and reflective in some areas and choppy and non-reflective in others.
- Generally you will find that normal (50 mm) to moderate telephoto (100 mm) focal lengths are easiest to work with. Most of my panoramas have been shot at 40-70 mm. These focal lengths capture scenics with the same perspective as the human eye. Remember, the wider the individual shot the less detail and information you are recording. Compare, for example, detail in a 20 mm shot with that for a 600 mm shot. The idea in a panorama is to capture a wide-angle composite view with a narrower angle-of-view lens.
- Even the best laid out panoramas require cropping which is a subtractive process. Be mindful of that by leaving enough room at the top and bottom of your sequence of photos to crop out some foreground and background in Photoshop. Furthermore, since the camera must be level, balancing foreground and background may require moving up or down in elevation. Sometimes this is possible. At other times you may find yourself walking away from a panoramic scene or deleting it after seeing it on the computer screen. This can be solved by using an omni-directional setup which requires an additional equipment expense.
- Having the sun directly behind you creates fairly even skies. Watch out for your shadow, however.
- If you capture an object that is moving, make sure it is in the middle of only one frame where there is no overlap with the other frames.
- Because the nodal slide is graduated, make note of the position and keep a list of nodal points for each lens/focal length so that you don’t have to repeat this time-consuming step.
- Panorama images can get quite large, so you may need more RAM in your computer. RAM is the cheapest way to increase the performance of your computer.
- Be creative and have fun. Try combining photo stitching with macros, extreme telephotos, high dynamic range, and even extended depth of field (e.g., focus stacking).
- Above all, learn how/when to break the conventional wisdom including all of the aforementioned tips and hints.
Resources
- “Get 4×5 Quality with a DSLR,” Dennis Frates, Outdoor Photographer, August 2009.
- This website provides a technical discussion about nodal points: http://www.vrphotography.com/data/pages/techtutorials/technotes/nodalptalign-tn.html.
- “The GigaScape,” George Lepp and Kathryn Lepp, Outdoor Photographer, November 2009.
- Really Right Stuff, Panoramas Made Simple, http://reallyrightstuff.com/WebsiteInfo.aspx?fc=108.
- Really Right Stuff, tabulated nodal points, http://reallyrightstuff.com/WebsiteInfo.aspx?fc=146.


















