Last week, I mentioned assembling multiple images into a single panoramic image. After a few people asked me about it, I thought I would elaborate on exactly how to make a panoramic image. The best way is to use a film camera specifically designed to make panoramic images. These cameras use 6X17 film instead of 35mm. However, since most of us don’t have cameras specifically designed for panos, let’s talk about how to make panos with what most of us do have.The process of combining multiple images into a single panoramic is called stitching. Think of it as stitching multiple images together like stitching patches of a quilt together. So the first question that comes to mind is why not just crop a single image into a panorama? Well, it’s all about maximizing the pixel real estate. Imagine if we had a digital image and we crop off the top and bottom thirds. We end up with only a third of the image left, and if we started with a nice large twelve megapixel file we are now left with only a 4 megapixel image. The alternative is to capture several images and lay them side to side in one bigger image. If we use five of the same 12 megapixel file we end up with an image that is nearly 60 megapixels in size. With a bigger file we have a lot more options including bigger prints.The first step is finding a subject that lends itself to a panoramic image. This is one of the hardest things about creating panos. The image should be interesting and it should lead the viewer from one side of the image to the other. The biggest pitfall newcomers to panoramic photography encounter is trying to cram too much into the image, and it ends up being cluttered and uninteresting.
The next step is in how you capture the image. Regardless of what type of camera you use, it is important to orient the camera perpendicular to the direction of the panoramic. For example if the long side of your panoramic is horizontal then orient your camera vertically in order to maximize your pixel real estate. Make sure to use a manual mode in your camera, and avoid using a polarizing filter so that all of the images have the same exposures. Lastly, try to avoid having both near and far elements in the image. The near object will appear to shift from one image to another. To see how this works, hold your thumb up at arm’s length and close one eye and then the other. You’ll notice how it appears to shift back and forth.
Notice how these images were captured with the camera oriented vertically without about 1/3 of each capture overlapping the previous one.
The final step is to combine the images you just took. Most digital cameras now come with software and instructions for how to create panoramic images. You can also do it manually in Photoshop. If you have Photoshop CS2 or CS3, you can select the images you want to combine in Bridge and then click, Tools>>Photoshop>>Photomerge in order to bring up the merge dialog. Select Auto or Perspective and make sure to select “leave as layers” if you are using CS2. CS3 automatically leaves them as layers. By leaving the images as layers, you can go back in and tweak the transitions to make them completely invisible.
Using Adobe Bridge for CS3, simply select all of the images you would like merged and the click Tools>>Photoshop>>Photomerge :
In the merge dialog, try selecting auto first. If the images lack enough features to allow Photoshop to guess on the arrangement, or if you are using an older version, you will have to choose a manual mode and align the images yourself:
The merge will take several minutes to complete depending on your computer. Once it is done, open up the layers pallette by clicking Window>>Layers or pressing F7. The black and white image next to each layer icon in the pallete is called a layer mask. Layer masks work by showing or hiding the content of a layer without affecting the actual image content of the layer. Clicking the little “eye” icon on the left of each layer will show or hide the layer:
Carefully examine the transitions between the images to look for spots that don’t match. By selecting a layer mask and using the brush tool, you can either paint with black to hide the layer, or paint with white to show more of it. Setting your brush opacity to about 30% will let you blend the layers together more evenly to hide any flaws:
The next step is to flatten the image. Although it may be a good idea to save a copy of the image as in, a flat image will be easier and faster to work with. Right click on the name of any layer in the layers pallette and the click “flatten.” Once you flatten the image you will be unable to go back and change the transitions, so make sure you are happy with them before flattening:
Unless your camera was miraculously straight, a feat I have yet to accomplish even with a level tripod, the outside edges will be uneven. Using the crop tool, select an area that will crop out all of the white background and press enter:
Viola! Now you have an image that has enough “pixel real estate” to print several feet wide. My final finishing touch was a simple black and white conversion:
For a more photography how-to tips, visit www.timothyfaust.com. Timothy Faust is an award winning photojournalist living in Breckenridge Colorado. If you have a photography question you would like to see answered in this column, please send it to email@example.com.