DocumentCode
3266576
Title
Wavelet coding for remote sensed data
Author
Rahman, Zia-ur
Author_Institution
Sci. & Technol. Corp., Hampton, VA, USA
fYear
1996
fDate
Mar/Apr 1996
Firstpage
452
Abstract
Space missions are increasingly requiring high resolution imagery and, hence, high data compression-two tasks which are inherently incompatible. A tradeoff must, therefore, be made between the available bandwidth and achievable resolution. An efficient visual communication channel-the process from image acquisition to image display-is key to the efficient transmission of the high resolution data. Our algorithm uses wavelet decomposition followed by entropy coding for compression as an integral part of the design of the visual communication channel, where the encoding process constitutes an additional noise source. This allows the development of analysis and synthesis filters which depend not only on the wavelet function, but also on the critical factors which constrain the visual communication channel itself. We feel that this approach is essential in designing a channel which provides the most information, and hence the highest resolution, for the least data
Keywords
data compression; image coding; image resolution; radiotelemetry; remote sensing; satellite telemetry; transform coding; visual communication; wavelet transforms; algorithm; analysis filter; available bandwidth; data compression; entropy coding; high resolution data; high resolution imagery; image acquisition; image display; noise source; remote sensed data; resolution; space missions; synthesis filter; visual communication channel; wavelet decomposition; Algorithm design and analysis; Bandwidth; Entropy coding; Filters; Image coding; Image resolution; Remote sensing; Space missions; Visual communication; Wavelet analysis;
fLanguage
English
Publisher
ieee
Conference_Titel
Data Compression Conference, 1996. DCC '96. Proceedings
Conference_Location
Snowbird, UT
ISSN
1068-0314
Print_ISBN
0-8186-7358-3
Type
conf
DOI
10.1109/DCC.1996.488384
Filename
488384
Link To Document