EM 1110-2-2907
1 October 2003
Figure 5-23. Landsat 7 Image of the Boston, Massachusetts area. Image on the
right shows red box outlining the boundaries of the subset scene on the right.
(3) Multiple-image Temporal Analysis. It is possible to combine bands
from different images or data sets. This allows a user to perform a change detec-
tion analysis. The process of "layering" multi-temporal data involves loading a
composite of bands from different images of the same scene. For example, a study
assessing urban development in a forested area would benefit from examining a
band combination that included band 3 data in the red plane, and band 3 data of a
later image in the green plane. If the spectral signature of the scene has changed
and is detectable within the resolution of the data, then changes in the scene will
be highlighted. This image can then be classified and the areas of change can be
statistically assessed. To perform this task accurately, it is important that both im-
ages are registered properly. Misregistration will lead to an offset in the image,
which leaves brightly colored lines of pixels. Be sure to choose images whose
data were collected under similar conditions, such as the same season, time of
day, and prevailing weather, i.e., minimum cloud cover.
f. Remote Sensing and Geospatial Information. Remote sensing data are eas-
ily integrated with other digital data, such as vector data used in a GIS (Geo-
graphical Information System). Vector data can be incorporated into a raster sat-
ellite image by overlaying the data onto an image scene. Conversely, a raster
image can be saved as a .jpeg or .tiff file and exported to a vector software proc-
essing program. Remote sensing data files can provide land cover and use infor-
mation as well as digital elevation models (DEMs), and a number of geo-physical
and biophysical parameters. Satellite images coupled with GIS data can be used to
create original maps. The use of remote sensing in this type of application can
drastically cut costs of GIS database development. It also provides data for inac-
cessible areas.
(1) Digital Orthoquadrangle (DOQs). A digital orthoquadrangle (DOQ) is
a digital image of an aerial photograph that has had ground relief removed and is
5-40
Previous Page
