Page 178 - Digital Analysis of Remotely Sensed Imagery
P. 178
CHAPTER 5
Image Geometric
Rectification
emotely sensed images obtainable from Earth resources sat-
ellites form a vital information source for managers of the
Renvironment and natural resources. Thanks to their ease of
acquisition and up-to-datedness, satellite images, together with
aerial photographs, are playing an increasingly significant role in
numerous geographic information system (GIS) applications. Unlike
existing maps, raw satellite images and scanned aerial photographs
have a local coordinate system. They do not have the right projection.
Neither do they have any scale nor a proper orientation. During data
acquisition the platform on which the sensor is mounted is in a state
of constant motion. Any deviation of the sensor position and orienta-
tion from the norm will lead to geometric distortions in the resultant
satellite images.
It is important to geometrically rectify these images for three reasons:
• Firstly, the end product of a large number of remote sensing
applications about the Earth’s resources and environment
occurs in the thematic map form. These maps must conform
to certain geometric mapping standards. Image rectification
ensures that geometric distortions inherent in the remote
sensing imagery are eliminated or reduced to an acceptable
level.
• Secondly, in order to be analyzed with data from other sources
in a GIS, raw satellite images and aerial photographs have to
be projected to a common ground reference system, which
enables images obtained at different times to be spatially
registered with one another.
• Finally, if remotely sensed data are used to detect changes or
update existing maps, they must be reprojected to a coordinate
system with a known geometry identical to that of the digital
maps to be revised.
143