DocumentCode :
3296122
Title :
Minimising the co-registration window size for SAR image pairs
Author :
Homer, J. ; Longstaff, I.D.
Author_Institution :
Dept. of Comput. Sci. & Electr. Eng., Queensland Univ., Brisbane, Qld., Australia
Volume :
5
fYear :
1998
fDate :
6-10 Jul 1998
Firstpage :
2345
Abstract :
Due to the presence of SAR speckle noise, co-registration of SAR image pairs typically involves local area rather than feature based matching. In some cases, e.g. airborne acquired SAR image pairs the local co-registration parameters change relatively quickly and nonsystematically across the imaged scene. Highly localised co-registration, that is a small local area window size is consequently required. Too small a window size however, can lead to inaccurate estimation of the co-registration parameters. This paper presents a theoretically based procedure for minimising the window size while retaining confidence in the co-registration accuracy. Results of an experimental study are included
Keywords :
geophysical signal processing; geophysical techniques; image registration; radar imaging; remote sensing by radar; synthetic aperture radar; terrain mapping; SAR image pair; SAR image pairs; accuracy; co-registration window size; geophysical measurement technique; image registration; land surface; local area matching; local co-registration parameters; minimisation; radar imaging; radar remote sensing; speckle noise; synthetic aperture radar; terrain mapping; window size; Computer science; Image sensors; Layout; Least squares methods; Parameter estimation; Radiofrequency interference; Rotation measurement; Size measurement; Speckle; Synthetic aperture radar;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Geoscience and Remote Sensing Symposium Proceedings, 1998. IGARSS '98. 1998 IEEE International
Conference_Location :
Seattle, WA
Print_ISBN :
0-7803-4403-0
Type :
conf
DOI :
10.1109/IGARSS.1998.702208
Filename :
702208
Link To Document :
بازگشت