Professional Documents
Culture Documents
In this section you will investigate constraints on the design of remote sensing systems.
A real-time data- In this system, the instant a signal over a terrain element was
handling system generated, it would be processed onto an interpretable format
and recognised as being unique to the particular terrain
element from which it came. This processing would be
performed nearly instantaneously (`real-time’), providing
timely information. Because of the consistent nature of the
energy/matter interactions, there would be no need for
reference data in the analysis procedure. the derived data
would provide insight into the physical - chemical - biological
state of each object of interest.
System The 'ideal' system
component
Multiple data users These people would have knowledge of great depth, both of
their respective disciplines and of remote sensing data
acquisition and analysis techniques. The same set of data
would be transformed into various forms of information for
different users. This information would be available to them
faster, at less expense, and for larger areas than information
collected in any other manner. With this information, the
various users would make profound, wise decisions about
how best to manage the earth resource under scrutiny, and
these strategic management decisions would be implemented
– to everyone’s delight !
2 BUT...
BUT there are problems encountered in the design and operation of real sensing
systems and it is obvious that an ideal remote sensing system does not and cannot
exist. Real remote sensing systems fall far short of the ideal at virtually every point
in the sequence outlined.
Let us consider some of the basic shortcomings common to all real remote sensing
systems in order to better understand their general operation and utility. The
following section is based on Lillesand and Kiefer (1994), 32-35.
● The energy source. All passive remote sensing systems rely on energy that
is either reflected and/or emitted from Earth surface features. As already
discussed, the spectral distribution of reflected sunlight and self-emitted
energy is far from uniform. Solar energy levels obviously vary with respect to
time and location, and different Earth surface materials emit energy to
varying degrees of efficiency.
While you have some control over the nature of sources of energy for active
systems, the sources of energy used in all real systems are generally non-
uniform with respect to wavelength and their properties vary with time and
location. Consequently, you normally must calibrate for source
characteristics on a mission-by-mission basis, or deal with relative energy
units sensed at any given time and location.
● The sensor. At this point, it should come as no surprise to you that an ideal
“supersensor” does not exist. No single sensor is sensitive to all
wavelengths and all real sensors have detectors with fixed limits of spectral
sensitivity. They also have a limit on how small an object on the Earth’s
surface can be and still be “seen” by a sensor as being separate from its
surroundings. This limit, called the spatial resolution of a sensor, is an
indication of how well a sensor can record spatial detail. Thus, depending on
the spatial resolution of the sensor and the heterogeneity of the ground area
being sensed, digital images comprise a range of “pure” and “mixed” pixels.
In general, the larger the percentage of mixed pixels (mixels), the more
limited is the ability to record and extract spatial detail in an image – and
empirical evidence suggests that most pixels on most images are mixels.
● The choice of a sensor for any given task always involves tradeoffs.
For example, photographic systems generally produce images of very fine
spatial resolution, but they lack the broad spectral sensitivity obtainable with
non-photographic systems. Many digital imaging systems are quite complex
optically, mechanically, and/or electronically and may have restrictive power,
space and stability requirements. These requirements often dictate the type
of platform from which a sensor can be operated. Platforms can vary from
stepladders to space stations. Depending on the sensor/platform
combination needed for a particular application, the acquisition of remote
sensing data can be a very expensive endeavour.
● The multiple data users. Central to the successful operation of any remote
sensing system are the users of the data. The “data” generated by remote
sensing procedures become “information” only if and when you understand
their acquisition, know how to interpret them, and know how best to use
them. A thorough understanding of the problem at hand is paramount to the
productive application of any remote sensing methodology. Also, no single
combination of data acquisition and analysis procedures will satisfy the
needs of all data users (Lillesand and Kiefer, 1994).