Home

Robert D Feinman

Combining Images in Photoshop



Original Scan

Here is the original scan of two panormaic images. They were taken with a Noblex
swing lens panoramic camera at Waimea Canyon on Kauai, Hawaii.
This produces about 125 degrees on a single image.
We shifted the camera for the second image so that we can combine them to get a
full 180 degree image. For a 360 degree image we usual shoot four pictures at 90 degree
intervals even though three would theoretically be enough. This give lots of overlap
to work with in the following steps.
We scanned them at one time so that the density and contrast would match.

Notice that the images are on the wrong sides and that there is extra black material.
These will be fixed in the next steps.

Layers Pallet
Here is the layers panel after we have selected the right image and cut it onto a new layer.
We have also shifted the images so that they are on the correct side and set the background
 to be its own layer.

Moved Images


Here is the image corresponding to the layers shown above.
They are on the correct side and most of the extra black border has been removed.

Layers Panel with masks
This is the layers panel after adding masks.
This is done by control clicking on the image thumbnail followed by the
add mask icon (rectangle with white circle) at the bottom of the panel.
This adds a mask filled with black wherever the image is missing.

Notice the highlighting around the mask icon on the top layer, this indicates the
further editing will be done on the mask and not the image. This is important to
keep track of later on.
Overlapped images

We now reduce the opacity of the top layer to about 50% and slide it over the lower image until some
important feature lines up
. In this case we used a notch in the horizon.
It may be necessary to shift and rotate the image if it doesn't quite match.
It is just necessary to get the images to line up at one vertical line.

masked image
We have now gone to the channels panel and clicked on the mask icon to make it visible.
By painting on the mask with a feathered brush we can soften the transition between
the two images. It is also possible to remove parts of the mask by painting with white.
While we are showing the mask as red it is easier to edit if the mask in not set to be visible.
It is important to make sure you are editing the mask and not the image.
Either make sure the mask is highlighted in the layers panel or is selected in the channels panel
with the eye turned off.

Final image
Here is the final image after being flattened.
The transition between the images can be softened by using
the clone or healing brush tools to copy pixels from one layer to
the other before flattening.
It has also been cropped. Sometimes if
the joins leave one image higher than the other it may be necessary to
clone in some image at the top or bottom to fill in the missing details if
the image is not to become too narrow.
The best practice is to keep the camera level from shot to shot.

Sponsored Advertisement

Back to Tips

Take some time to enjoy the rest of the site
Send email to robert.feinman@gmail.com

© 2002 Robert D Feinman