DocumentCode :
2250462
Title :
ADPCM for advanced LANDSAT downlink applications
Author :
Brower, B.V. ; Couwenhoven, D. ; Gandhi, B. ; Smith, C.
Author_Institution :
Commercial & Gov. Syst. Div., Eastman Kodak Co., Rochester, NY, USA
fYear :
1993
fDate :
1-3 Nov 1993
Firstpage :
1338
Abstract :
Under funding from the Defense LANDSAT Program Office, a method was developed for compressing multispectral data. A rate-controlled adaptive differential pulse code modulation technique was developed with minimal complexity for downlink applications. This algorithm uses an adaptive 2-D and 3-D prediction of pixels. The difference between the predicted and original pixel is quantized with a locally adaptive quantizer. This technique has produced compression ratios of up to 2.5:1 losslessly and up to 5:1 with minimal visual image quality loss on LANDSAT and M-7 high resolution multispectral data. This paper describes the ADPCM algorithm and its impact on numerical, visual and machine exploitation performance
Keywords :
analogue-digital conversion; data compression; image coding; pulse-code modulation; remote sensing; satellite relay systems; ADPCM algorithm; Defense LANDSAT Program Office; LANDSAT downlink applications; adaptive 2-D prediction; adaptive 3-D prediction; adaptive differential pulse code modulation; adaptive quantizer; compression ratios; high resolution multispectral data; image coding; machine exploitation performance; multispectral data compression; numerical performance; rate controlled ADPCM; visual image quality loss; visual performance; Downlink; Image coding; Image quality; Image resolution; Modulation coding; Multispectral imaging; Pixel; Pulse modulation; Remote sensing; Satellites; Spatial resolution;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signals, Systems and Computers, 1993. 1993 Conference Record of The Twenty-Seventh Asilomar Conference on
Conference_Location :
Pacific Grove, CA
ISSN :
1058-6393
Print_ISBN :
0-8186-4120-7
Type :
conf
DOI :
10.1109/ACSSC.1993.342323
Filename :
342323
Link To Document :
بازگشت