Digital Images and Seismic Design
A new role for photographs in architecture and engineering
AIA Faculty Institute on Teaching Seismic Design
Berkeley CA, February 2-5, 1996
Last updated Thursday, February 1, 1996, at 5:09 PM
Copyright © 1996,
Kirk Martini
Table of Contents
Overview
Digital images offer several advantages over conventional
slides in for teaching and research in architecture and engineering. These advantages include:
- Availability: There are now sources where images are available only
in digital format, and cannot be obtained as conventional slides. Particularly
important are sources such as the EqIIS system of the EERC Library which provides
images free in the public domain.
- Enhancement: Using image processing software, it is possible to
enhance the contrast, color, and sharpness of a digital image for greater
clarity. It is also possible to distort the fidelity to clarify important
engineering features, such as digitally "bleaching" a dark slide to reveal
details in the shadows. In addition, images can be cropped and manipulated
to remove extraneous information.
- Annotation: With digital images, it is possible to add text, symbols,
transparent overlays, guidelines, and other annotation to highlight, explain,
and analyze important features in an image.
- Distribution: Unlike conventional slides, digital images can be
distributed over the world wide web, so that they can be made accessible to
any computer in the world that is connected to the web.
- Integration with text: Using web documents, digital images
can be easily combined with sketches and text which support and explain the image
content. This document demonstrates this capability.
Availability
The EqIIS collection of the EERC Library is by far the best source
of seismic and structural images on the web. FEMA maintains a very small
site, although some of the images are instructive.
After the Kobe and Northridge events, many sites sprang up from news organizations
and ordinary people with scanners and servers. The level of image quality and
documentation for these sites is generally not very good, and many of the sites
were short lived. There can also be
copyright difficulties in using images from news organizations.
Here are some sample links:
Enhancement
There are several useful techniques for digital enhancement of
images. The most important techniques are adjusting levels and
sharpening. Cropping an image is also useful to remove extraneous
information on the edges of an image, it is also possible, though
not always ethical, to actually remove elements from a scene. The
following is a series of examples illustrating these techniques.
Adjusting Levels
The thumbnail image below is linked to a downloaded page from the
EqIIS Collection. Click on the image to see the information page,
and then click on that page's thumbnail to see the full-sized
image.
The effect of the weld damage on the paint is very difficult to see
in the image because the slide is so dark.
Compare that image to an image which has been enhanced by adjusting
the exposure levels, shown in this comparison.
This example illustrates the ability to distort visual fidelity to clarify
engineering features, and it raises an important issue concerning image
enhancement. Effective image enhancement must address the question of how
the image will be used. The enhanced image of the weld is very useful for
identifying locations where the paint has flaked, but is may be much less
useful for training someone to recognize such damage. Enhanced images can
be used to represent situations that in fact never existed, and must be used
with care.
Sharpening
The following comparison
shows an image of the Olive View Hospital, downloaded
from the EqIIS collection, with a version of the same image
that has been sharpened, cropped, and scaled, along with a side-by-side
comparison of portions of the images.
Cropping and scaling allow the image to focus on the primary
subject, removing the landscaping and scenery in the foreground
and periphery. Sharpening simply creates a clearer, better looking
image; this can be important in images with fine detail.
Annotation
Digital images also allow the annotation of images with text,
guidelines, overlays, and other graphic devices. Such annotation
is not impossible with conventional images, but is extremely
time consuming and expensive. With digital images, it is possible
to create rich annotations working with a desktop computer.
The series of thumbnail images below use an image of an earthquake-damaged
column to explain the concept of shear.
In addition to annotation with symbols, as illustrated with the force
arrows, these images illustrate two useful techniques:
- Masking: The gray mask in the second image helps to isolate the
area in question. For this image, it identifies the free body on which the
force arrows are acting in the third image. Without masking it would not be clear
whether the arrows were intended to represent forces acting on the object or forces exerted
by it.
- Overlays: Semi-transparent overlays, as used in the third image, are
useful for several purposes. As used above, the overlay clarifies the racking action
resulting from shear, and allows the reader to make a connection between the mathematical
abstraction of shear deformation and the behavior of real structures. Overlays are also
useful for keying an area of an image to associated text.
The series of images below uses these techniques to illustrate the concept
of bending moment.
Distribution
Conventional 35mm slides are usually only viewable for a few minutes
in a lecture hall or classroom. The reproduction costs of color
photographs preclude wide distribution. Digital images can be distributed
over the world wide web, and are thus accessible to any computer in the
world connected to the web. In teaching, this means that assignments can
be based on viewing and analyzing the phenomena in images, and changes the
role that photographs can play. The combination of image annotation and
outside access makes it possible to create a much tighter linkage of
structural theory and structural behavior.
Integration with text
As demonstrated in this document, digital images on the web
can be integrated with supporting text to create a complete
analysis and presentation of an idea. This medium makes it
possible to effectively publish a slide talk, where the
images serve as slides, the text serves as speech, and
annotation serves as a laser pointer. The thumbnail image
below is linked to a page containing an untouched EqIIS image,
followed by a version of that image that has been cropped, scaled,
sharpened, annotated, and accompanied by supporting text.
Slide talks are often the centerpiece of conference
presentations, but they are never published in complete form
because of the expense of color reproduction. Digital
imaging and the web offer the possibility of enriching
education and research by giving photographs a much more
integrated role in the exchange of ideas among architects, engineers,
and the host of other disciplines involved in seismic design.
Last updated Thursday, February 1, 1996, at 5:09 PM
Copyright © 1996,
Kirk Martini
Please send comments or questions to
Martini@virginia.edu