Posted By Timothy J. Bennett CRA, OCT-C,
Friday, May 10, 2013
Updated: Friday, May 10, 2013
A flatbed scanner is often used as a small
object camera for flat subjects. Surprisingly, they are also capable of
imaging small three-dimensional objects as well. The optics in a
flatbed scanner are designed to record the entire 8.5 x 11 (or larger)
scanning bed on a small sensor. In order to do that, the scanner
utilizes a wide angle lens with mirrors or prisms to place the sensor at
an optical distance much greater than the actual physical distance of
the shallow scanning bed.
Although the plane of focus is set at
the surface of the glass plate, the optics provide better than expected
depth-of-field, enabling a reasonable degree of sharpness in small
three-dimensional subjects imaged on the glass bed.
The optics also introduce a certain
amount of parallax when a three-dimensional object is moved to different
lateral positions on the scanning bed.
You can see the effect here with this
film box placed in left, center, and right positions on the bed. This
parallax allows for generation of sequential stereo pairs much like the
stereo pairs we create by shifting the fundus camera laterally between
two sequential photos.
With standard stereo photography, the
typical camera shift distance between photos is 2.5 inches which
approximates the average distance between eyes. Another commonly used
convention is to shift the camera 1/30th of the distance to
the nearest point in the scene. The optical properties of a scanner
render these common conventions irrelevant. I suggest experimenting with
your particular scanner to judge a good distance between images but
I've found that one to two inches works well, and sometimes as little as
half an inch produces enough parallax to achieve a stereo effect.
Alignment between successive scans of
unevenly shaped objects can be challenging. Objects can roll or shift
between shots, distorting the stereo effect. If necessary, objects can
be attached to a firm background with tape or glue. A glass front box
can also be used to hold the object on the scanner. In both of these
cases, the glass box or background is easily moved side-to-side on the
scanning bed between shots to create parallax while maintaining the same
The optics of the scanner are usually
sharpest near the center of the scanner bed, so try to place objects
near the middle. Judicious use of a Sharpen or Unsharp Mask
filter in your photo editing software may help as well. When aligning
the stereo pairs for viewing, any vertical displacement must be avoided.
Leave some empty space around the object to allow for cropping any
excess vertical overlap.
To achieve a dark background, you can
simply leave the scanning lid open and turn off any ambient room light.
Placing the scanner lid over the object will result in a lighter
background that varies with the distance of the lid from the bed.
To prevent the lid from pressing against
fragile objects or accidentally moving the object during lid placement
between successive images, small blocks of foam board can be used to
prop the lid above the object.
Several years ago, I read of a
technique for scanning three-dimensional objects by placing them in a
shallow box and inverting the scanner on top of the box. The objects
shown above were placed in a shallow box lid and imaged with this
technique, but I don't recommend it. Placing the scanner upside down can
wreak havoc with the gears in the scanning carriage or stepping motor.
Once you've captured stereo pairs, they
can be cropped, aligned, and rendered as anaglyph images in photo
editing software or in dedicated stereo alignment software such as
Stereo Photo Maker.
Give this technique a try and have fun!
The following anaglyph images were all
done on a flatbed scanner. To view these images in stereo, use red/cyan
(blue-green) anaglyph glasses with the red lens covering the left eye.
To see an anaglyph video clip panning across this image, click here
Timothy J. Bennett, CRA, FOPS, OCT-C,
is an ophthalmic photographer in the Penn State University Department of
Ophthalmology at the Milton S. Hershey Medical Center. Mr. Bennett has
over 30 years experience working in university-based academic medical
centers. He is a nationally recognized author, lecturer and educator in
the field of ophthalmic photography. Areas of expertise include
fluorescein angiography, digital imaging, monochromatic photography,
fundus autofluorescence and optical coherence tomography.
Mr. Bennett holds current certification
as a Certified Retinal Angiographer and Optical Coherence Tomographer,
is a former Commissioner of the Joint Commission on Allied Health
Personnel in Ophthalmology, and has been recognized for his
contributions to the profession by being named a Fellow of the
Ophthalmic Photographers Society. He has served on the OPS Board of
Certification, the OPS Board of Directors, and is Past-President of the
Ophthalmic Photographers' Society.