The old and original model of creating a username and password for every web site is being strained under the growing popularity of the Internet and the number of web sites people need to “authenticate” themselves to. The Internet needs a new authentication scheme, and “OpenID” may just be it. With OpenID, users authenticate with a SINGLE identification and any other OpenID-supporting sites are able to trust that. Leo and I will explain how all that works.

Let’s look at a few techniques and tips about how to create great panoramas
simply and with a relatively low-cost approach.

Panoramas have always fascinated people. Some of the early approaches to
creating panoramas involved cameras with very wide lenses, or lenses that
moved and painted the image on the film. In fact, high-end panoramas are
still done this way.

These days, with digital photography, creating panoramas is much simpler.
And the latest software automates the stitching process of blending a series
of images together to create a much bigger single image.

Of course, there is an enormous range of software available from around $30
to $1,000 - even free programs. Some programs do the stitching
automatically, and even balance exposure and align pictures that are
lessthan perfect. Others provide complicated sets of tools with steep
learning curves to allow for full manual alignment and stitching.

We’re going to look at the automatic tools, and emphasize the art of taking
the pictures so that the images are at their best for these tools to work
with.

Why create a panorama?

- To create a total sense of horizontal expanse, such as a sweeping view
- To capture a very wide subject when you can’t get back far enough
- To show a big picture that has close-up details
- To create a sense of being there, such as with a street scene
- To capture a very tall subject, with a vertical panorama
- Because it’s fun and a really cool thing to do

Key factors when capturing a panorama:

- Equal exposure on each image that is part of the panorama (especially the
sky)
- The correct amount of overlap, between 15% and 50% depending on the
subject
- Keeping the horizon horizontal and at the same level in the picture
- Using the optimal focal length for the image to avoid distortion, yet
provide coverage
- Keeping the camera on the same axis for all the shots

Choose subjects that will work well as a panorama:

- Avoid extremes of lighting from one side to the other
- Look for interesting elements to create a subject or anchor
- Check from left to right that everything will fit in each frame while
keeping the same horizontal plane

Key things you need to do when you’re taking a panorama:

1. Use a tripod if you can. However, in my experience, the most inspiring
panoramas have presented themselves when I don’t have a tripod.
2. Choose a focal length that is the equivalent of a normal lens or even a
little more telephoto. Wide-angle lenses can create distortions in
perspective, and are more likely to give you vignetting, or dark corners
which are a nightmare to correct later. If you can’t get the top and bottom
parts in, turn your camera to a vertical format and take overlapping shots
that way.
3. If your final picture contains both near and far objects, you’ll need to
use a small aperture so that both the close objects and the horizon will be
in focus.
4. If your entire panorama is at a distance, and you’re using telephoto and
hand-holding, you’re better opting for a fast shutter speed.
5. Decide if you want to shoot in a jpeg or RAW. Jpeg will give you a small
file, and one that is pre-optimized and easier to stitch later. Choosing RAW
allows you more control over exposure later, but also gives you larger
files; which may be a challenge if you have a slow machine when it comes to
stitching.
6. Determine the optimal area of the final picture that you want to have
correctly exposed and in focus.
7. Lock your focus and exposure on this area for all the shots.
8. Position your body so you are facing the final shot, at the right hand
end of the panorama.
9. Keep your feet still and twist your body back to the point where you will
start. Ensure you have extra top and bottom for cropping.
10. Take the first picture, allowing about 30% more on the left-hand side
than you will need in the picture.
11. Take a note of some of the landmarks or key points in your image towards
the right hand side, that will guide you in the next step of overlap.
12. Keeping the horizon level, and at the same height, move to the next
image, an overlap around 20 percent. Use the key points that you remember
from the previous shot. Again, take a note of some key points on the right
hand side.
13. Continue this until you have completed the entire panorama, with a
little bit of spare subject on the right hand side of the last image.

* GOOD TIP: Make sure that nothing really important is at the very top or

very bottom of the picture. When you stitch these, you often lose areas at
the top and bottom when the software corrects for perspective.

Preparing to stitch the images to create a single panorama

First, decide whether your panorama is for screen viewing on a computer, or
whether you’ll be making a print with it. If it for screen viewing, decide
whether you want the entire panorama to show on a single screen, or do you
want a panorama that the viewer will need to scroll across. If you’re making
a print, what size would your final print be?

Based on the megapixels of your camera, will the final image have sufficient
resolution to create the print as is, or will you need to boost the size?
Generally speaking, it is best to boost the size once you have finished the
stitching and final corrections. For boosting image size, I typically use
Genuine Fractals from On One software (www.ononesoftware.com). Adobe
Photoshop CS3 (www.adobe.com) also has sophisticated algorithms for boosting
image size.

If your subject was very dark, and you boosted the ISO rating on your
camera, you’re likely to get more digital noise. Or if you’re using a
camera with a sensor that creates a lot of noise, then you’ll also need to
de-noise it.

The best time to de-noise is before you do anything else to the image. Most
software image programs like Photoshop have a de-noising function. I prefer
to use a third party product such as Noise Ninja from PictureCode
(www.picturecode.com). Be careful not to overdo noise reduction, or you’ll
lose details, and end up with an image that looks plastic.

Choosing your panorama software

The three that I will talk about today are Photoshop CS3 with its Photomerge
function, and two lower-cost options - DoubleTake (www.echoone.com) and
Calico from Kekus (www.kekus.com). These are both in the $20 to $30 range,
and do an excellent, stand-alone job with most images. Even when the images
are not perfect, both of these have their strengths for correcting
variations.

As with any work involving on-screen display, or print output, it is vital
that you have your display calibrated and color-corrected.

Now for the stitching

If any of your images are less than perfect for the panorama, it is better
to correct them before you start stitching. Things to correct include:
making sure your horizon is horizontal, correcting any lens distortions
(including vignetting) and equalizing the exposure if you forgot to lock it
went taking the shots. The standalone application, or Photoshop plugin,
LensFixCI from Kekus, is a very handy and quick way of automatically
correcting your image for the particular make and model of lens that you
used for each shot. Do this after noise reduction, but before loading into
the panorama software.

Also, if some of the shots have areas that are washed out, or deep shadow
areas, you may want to try working on these before you start stitching.
There are no absolute rules, and often I will do finishing touches in the
final assembled image. The hardest thing to fix once the panorama is made is
the sky. This is why it’s important to lock the exposure, so the sky is
equally exposed across the whole panorama.

You can either drag all your images to a window, as with Calico and
DoubleTake, or add them as files or a folder through a browsing dialog box
in Photoshop: File>Automate>Photomerge.

DoubleTake has a number of correction and adjustment controls built in.
These allow you to rotate and line up your horizon, adjust exposure,
finetune the placement of images for better alignment of points, and create
an optimal crop when you finish. DoubleTake allows you to save your file as
TIFF, PNG, PDF, QuickTime VR and jpeg, but not as Photoshop. Where images
have exposure variations, or noticeable vignetting, DoubleTake is not as
strong as the automatic fix in Calico.

Calico offers fewer image adjustment controls, is a better speed choice when
images are more correct to start with, and seems to do a better job of
combining images that have exposure variations and vignetting. It allows you
to save a file as TIFF, jpeg and Photoshop with layers (for further
adjustments), as well as choosing your method of interpolation, and exposure
equalization. Calico has the ability to stitch multiple rows of images.
Kekus, the developer of Calico, also offers other sophisticated tools for
manual creation.

Photomerge in Photoshop offers a number of different stitching and blending
options. For most purposes, the automatic function is the one used. It is
simpler, does a better job, and requires no fixing. Photoshop creates a
panorama with a combination of layers and layer masks. It actually chops the
images and dovetails them together after aligning a selection of location
points. If you try and move the layers after the stitching has occurred,
they will look as if they are being pulled apart like a cookie being broken.
But you can still work with the layers and masks for fine-tuning.

When you save your panorama, you may be offered the choice of saving with
layers, or in different formats. If you have the “Save As Layers” option, I
would probably do that and then “Save As” that version often as a flattened
Photoshop version for printing. Remember when you save as a jpeg, you do
lose some quality in detail each time you save as.

You now have a Photoshop file which you treat as you would any other image
file for printing. The difference is that you need a sheet or roll of paper
that is long enough to accommodate the width of the panorama. You may find
that some printers have a limitation on the width of panorama you can print
(meaning the length of the paper it will feed through).

Each printer maker has different setups for panoramas, so you need to make
sure you have the proper driver installed and follow the instructions
carefully. Not all manufacturers write instructions that make it easy to
print a panorama. However, if you persevere, the end results will be worth
it.

For large panoramas, you may want to send your file to a lab. A Toronto
firm, Silvano Imaging (www.silvanoimaging.com), is about to offer a service
that will print panos up to 51″ x 150′0″ (that’s 130cm x 45metres). Try and
e-mail THAT file (they have an FTP server, or you can send on a disk).

Finally, practicing your new skill

Go and play with some local pano-friendly subjects. Get some practice in now
so you’ll be ready when that amazing panorama opportunity presents itself in
the future!
—
Jim Everett
THINK180™

RELATED WEBSITE LINKS
www.itookthat.com – photography blog site with links
www.jamesgordongallery.com – online gallery

PRODUCTS SHOWN *Essential to show in segment – others only mentioned

* Adobe - Photoshop CS3 $649, Elements $99 - www.adobe.com
* Kekus - Calico Panorama $39, LensFix $25 - www.kekus.com
* Pantone – Huey $80 Pro $120 – www.pantone.com
* PictureCode - Noise Ninja $45 Pro $80 - www.picturecode.com

Silvano Color Labs Toronto - large format printing, canvas www.silvanoimaging.com
Epson USA - Stylus Photo Printers and Papers - http://www.epson.com
Canon – PhotoStitch (free with Canon cameras) - www.usa.canon.com
Nikon - D200 SLR
Apple iLife08 - iPhoto - www.apple.com

The most common use of a combination chart is to show two series that are a different order of magnitude. To show Revenue in Millions and profit percentage on the same chart, follow these steps:
Create a chart with both revenue and profit. You won’t be able to see profit.
Choose Series Profit dropdown
Move to the Secondary Y-axis in the Format Series dialog. The problem is that now the columns are directly on top of each other, making it impossible to see the smaller numbers.
Change the chart type of the profit series from column to a line.

A completely different use for combining chart types is to draw in custom gridlines.
Create a chart showing some data
Type a new data range with X=Custom gridline locations and Y=0. Add this series to the chart. It completely ruins the first chart. That’s OK.
Change the series to an XY chart.
Make the markers be transparent
Add an Error Bar extending to the right. This becomes the Gridline!

Show an example of using this teqhnique to create four stacked surface charts.

* I will post show notes at http://www.mrexcel.com/tip149.shtml