Page 108 - Digital Analysis of Remotely Sensed Imagery

Overview of Remotely Sensed Data 79

2.8.1 Identification of User Needs
Different users purchase remote sensing data for different purposes
and needs. Before deciding what type of data to purchase, the analyst
needs to identify the special requirements by answering the following
questions:

• First, what type of data is the most useful for studying the
phenomenon in question? Is it optical or microwave? If
optical, should the data be recorded in visible light, NIR,
middle infrared, far infrared, or their combination? Optical
data are the best at studying in-water constituents, while
near and middle infrared imagery is especially good at
studying vegetation. TIR imagery is effective at revealing
heat-related phenomena, but suffers from a lack of detail
and coarse spectral resolution. Microwave imagery is
excellent at detecting hidden features and at ocean
applications. It is the only remote sensing functioning in
tropical areas where the ground area is frequently obstructed
by clouds. Radar imagery suffers from a lack of spectral
resolution and radiometric noises. For most natural
resources mapping and environmental monitoring, the best
choice is multispectral data over the visible and NIR portion
of the spectrum.
• Second, what detail level is required? The amount of details
to be identified from remote sensing imagery is related
directly to its spatial resolution and scale. Images of varying
spatial resolutions enable different information to be derived
from them at different accuracy levels. Images of a finer
spatial resolution allow more details to be discerned, but they
may cover only a small strip of the area under study. Thus, a
large number of images have to be acquired to completely
cover the area. Another implication of using fine spatial
resolution images is the long time and huge cost necessary to
process them. Besides, data of a finer spatial resolution are
more expensive than those of a coarser resolution, so it is
important to select images at the right spatial resolution. In
deciding which image is the best for the task, users need to
base the decision on their special needs. It may not bring
much benefit if detailed information is not required. Apart
from the amount of details visible, spatial resolution or pixel
size also affects the reliability of the results derived from the
data. Covering a local area, images of a finer spatial resolution
are suited to applications of a small area as the mapping
accuracy requirements are stringent. The expected accuracy
standard can be met more easily with the use of images of a
finer spatial resolution. Coarser resolution data are more
appropriate for broad-scale applications. Geometric accuracy