Vous êtes sur la page 1sur 8

Question 1: The Process of image classification in satellite remote sensing

Many image processing and analysis techniques have been developed to aid the interpretation of remote
sensing images and to extract as much information as possible from the images. The choice of specific
techniques or algorithms to use depends on the goals of each individual project. In this section, we will
examine some procedures commonly used in analyzing / interpreting remote sensing images.
Pre-Processing
Prior to data analysis, initial processing on the raw data is usually carried out to correct for any distortion
due to the characteristics of the imaging system and imaging conditions. Depending on the user's
requirement, some standard correction procedures may be carried out by the ground station operators
before the data is delivered to the end-user. These procedures include radiometric correction to correct
for uneven sensor response over the whole image and geometric correction to correct for geometric
distortion due to Earth's rotation and other imaging conditions (such as oblique viewing). The image may
also be transformed to conform to a specific map projection system. Furthermore, if accurate
geographical location of an area on the image needs to be known, ground control points (GCP's) are used
to register the image to a precise map (geo-referencing).
(Reference: Dr. S. C. Liew)

Question 2: The difference between unsupervised classification and supervised classification


UNSUPERVISED CLASSIFICATION

SUPERVISED CLASSIFICATION

Pixels are grouped based on the reflectance properties

The user selects representative samples for each

of pixels. These groupings are called clusters. The

land cover class in the digital image. These sample

user identifies the number of clusters to generate and

land cover classes are called training sites. The

which bands to use. With this information, the image

image classification software uses the training

classification software generates clusters. There are

sites to identify the land cover classes in the entire

different image clustering algorithms such as K-means

image.

and ISODATA.

Unsupervised Classification Example

Supervised Classification Example: IKONOS

The user manually identifies each cluster with land The classification of land cover is based on the
cover classes. Its often the case that multiple clusters spectral signature defined in the training set. The
represent a single land cover class. The user merges digital image classification software determines
clusters into a land cover type. The unsupervised each class on what it resembles most in the training
classification

image

classification

technique

is set.

The

common

supervised

classification

commonly used when no sample sites exist.

algorithms are maximum likelihood and minimum-

Unsupervised Classification Steps:

distance classification.

Generate clusters

Supervised Classification Steps:

Assign classes

Select training areas

Generate signature file

Classify

Unsupervised Classification Diagram

Supervised Classification Diagram

(Reference: Blaschke T, 2010. Object based image analysis for remote sensing. ISPRS Journal of
Photogrammetry and Remote Sensing 65 (2010) 216)

Question 3: Sun-Synchronous orbit and they important of land resource Mapping


SUN-SYNCHRONOUS

Is a geocentric orbit which combines altitude

GEOSTATIONARY ORBIT

Altitudes of approximately 36,000 kilometers.

and inclination in such a way that an object on


that orbit ascends or descends over any given
point of the Earth's surface at the same local
mean solar time.

They cover each area of the world at a constant

Revolve at speeds which match the rotation of

local time of day called local sun time.

the Earth so they seem stationary relative to


the Earth's surface.

Travels northwards on one side of the Earth and

Allows the satellites to observe and collect

then toward the southern pole on the second

information continuously over specific areas.

half of its orbit.

The sensor "sees" a certain portion of the

Weather

Earth's surface and this refers to swath.

commonly have these types of orbits.

This apparent movement allows the satellite

Due to their high altitude, some geostationary

swath to cover a new area with each

weather satellites can monitor weather and

consecutive pass.

cloud patterns covering an entire hemisphere

and

of the Earth.

communications

satellites

Question 4: Describe geostationary orbit and when is this type of orbit used?
One very popular orbit format is the geostationary satellite orbit. The geostationary orbit is used by many
applications including direct broadcast as well as communications or relay systems. The geostationary
orbit has the advantage that the satellite remains in the same position throughout the day, and antennas
can be directed towards the satellite and remain on track. This factor is of particular importance for
applications such as direct broadcast TV where changing directions for the antenna would not be
practicable. It is necessary to take care over the use of the abbreviations for geostationary orbit. Both GEO
and GSO are seen, and both also used for geosynchronous orbit.

Arthur C Clarke's Geostationary Orbiting Satellites Concept

Geostationary orbit basics


As the height of a satellite increases, so the time for the satellite to orbit increases. At a height of 35790
km, it takes 24 hours for the satellite to orbit. This type of orbit is known as a geosynchronous orbit, i.e. it
is synchronized with the Earth.
One particular form of geosynchronous orbit is known as a geostationary orbit. In this type of orbit the
satellite rotates in the same direction as the rotation of the Earth and has an approximate 24 hour period.
This means that it revolves at the same angular velocity as the Earth and in the same direction and
therefore remains in the same position relative to the Earth.
In order to ensure that the satellite rotates at exactly the same speed as the Earth, it is necessary to clarify
exactly what the time is for the rotation of the Earth. For most timekeeping applications, the Earth's
rotation is measured relative to the Sun's mean position, and the rotation of the earth combined with the

rotation around the Sun provide the length of time for a day. However this is not the exact rotation that
we are interested in to give a geostationary orbit - the time required is just that for one rotation. This time
period is known as a sidereal day and it is 23 hours 56 minutes and 4 seconds long.
Geometry dictates that the only way in which an orbit that rotates once per day can remain over exactly
the same spot on the Earth's surface is that it moves in the same direction as the earth's rotation. Also it
must not move north or south for any of its orbit. This can only occur if it remains over the equator.

Geostationary orbit can only be over the Equator

Different orbits can be seen from the diagram. As all orbital planes need to pass through the geo-centre
of the Earth, the two options available are shown. Even if both orbits rotate at the same speed as the
Earth, the one labelled geosynchronous will move north of the equator for part of the day, and below for
the other half - it will not be stationary. For a satellite to be stationary, it must be above the Equator.
Geostationary satellite drift
Even when satellites are placed into a geostationary orbit, there are several forces that can act on it to
change its position slowly over time.
Factors including the earth's elliptical shape, the pull of the Sun and Moon and others act to increase the
satellite orbital inclination. In particular the non-circular shape of the of the Earth around the Equator
tends to draw the satellites towards two stable equilibrium points, one above the Indian Ocean and the
other very roughly around the other side of the World.. This results in what is termed as an east-west
liberation or movement back and forth.

To overcome these movements, fuel is carried by the satellites to enable them to carry out "stationkeeping" where the satellite is returned to its desired position. The period between station-keeping
manoeuvres is determined by the allowable tolerance on the satellite which is mainly determined by the
ground antenna beam width. This will mean that no re-adjustment of the antennas is required.
Often the useful life of a satellite is determined by the time for which fuel will allow the station-keeping
to be undertaken. Often this will be several years. After this the satellite can drift towards one of the two
equilibrium points, and possibly re-enter the Earth's atmosphere. The preferred option is for the satellites
to utilise some last fuel to lift them into a higher and increasing orbit to prevent them from interfering
with other satellites.
Geostationary orbit coverage
A single geostationary satellite obviously cannot provide complete global coverage. However, a single
geostationary satellite can see approximately 42% of the Earth's surface with coverage falling off towards
the satellite is not able to "see" the surface. This occurs around the equator and also towards the Polar
Regions.

Geostationary satellite coverage


For a constellation of three satellites equally spaced around the globe, it is possible to provide complete
coverage around the equator and up to latitudes of 81 both north and south.
The lack of polar coverage is not a problem for most users, although where polar coverage is needed,
satellites using other forms of orbit are needed.

Geostationary orbit and path length / delay


One of the issues with using satellites in a geostationary orbit is the delay introduced by the path length.
The path length to any geostationary satellite is a minimum of 22300 miles. This assumes that the user is
directly underneath the satellite to provide the shortest path length. In reality the user is unlikely to be in
this position and the path length will be longer.
Assuming the shortest path length, this gives a single trip i.e. to the satellite or back of a minimum of
around 120 mile-seconds. This means that the round trip from the ground to the satellite and back is
roughly a quarter of a second.
Therefore to obtain a response in a conversation can take half a second as the signal must pass through
the satellite twice - once on the outward journey to the remote listener, and then again with the response.
This delay can make telephone conversations rather difficult when satellite links are used. It can also be
seen when news reporters as using satellite links. When asked a question from the broadcasters studio,
the reporter appears to take some time to answer. This delay is the reason why many long distance links
use cables rather than satellites as the delays incurred are far less.

WEB REFERENCE
1. http://gisgeography.com/image-classification-techniques-remote-sensing/
2. http://www.crisp.nus.edu.sg/~research/tutorial/process.htm
3. http://www.grsgis.com/image-processing.html
4. https://en.wikipedia.org/wiki/Remote_sensing
5. http://www.oneonta.edu/faculty/baumanpr/geosat2/RS%20History%20II/RS-History-Part2.html
6. http://www.thefreedictionary.com/geostationary+orbit
7. http://en.wikipedia.org/wiki/Geosynchronous_orbit
8. http://en.wikipedia.org/wiki/Sun-synchronous_orbit
9. https://en.wikipedia.org/wiki/Geostationary_orbit#References
10. http://www.radio-electronics.com/info/satellite/satellite-orbits/geostationary-earth-orbit.php