DocumentCode :
3396035
Title :
An improved InSAR image registration algorithm
Author :
Sun, Wenchang ; Song, Jianshe ; Gu, Xiaofei ; Zhang, Lin
Author_Institution :
Research Institute of Information and Engineering, Xi´an Research Institute of High-tech, China
Volume :
2
fYear :
2010
fDate :
30-31 May 2010
Firstpage :
522
Lastpage :
525
Abstract :
A novel method is proposed as coarse coregistration, part of InSAR image registration. Edge detection of sar image is processed and control point selection is done subsequently. Then, control points selected in two images are matched and filtered using R-RANSAC method, so as to form tie points. From the tie points, warp function parameters are estimated, and coarse coregistration is finished. Based on coarse coregistration, fine coregistration can be carried out by performing correlation method on oversampled images. Experiments prove that the method is suitable for InSAR, and may enhance the coregistration procedure.
Keywords :
Correlation; Detectors; Image edge detection; Image registration; Interference; Matched filters; Object detection; Parameter estimation; Process control; Synthetic aperture radar interferometry; InSAR; R-RANSAC; corner detection; edge detection; image registration;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Industrial Mechatronics and Automation (ICIMA), 2010 2nd International Conference on
Conference_Location :
Wuhan, China
Print_ISBN :
978-1-4244-7653-4
Type :
conf
DOI :
10.1109/ICINDMA.2010.5538256
Filename :
5538256
Link To Document :
بازگشت