A region-growing algorithm for InSAR phase unwrapping

被引:226
作者
Xu, W [1 ]
Cumming, I
机构
[1] MacDonald Dettwiler & Associates Ltd, Richmond, BC V6V 2J3, Canada
[2] Univ British Columbia, Dept Elect & Comp Engn, Vancouver, BC V6T 1Z4, Canada
来源
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 1999年 / 37卷 / 01期
基金
加拿大自然科学与工程研究理事会;
关键词
digital elevation model; interferometry; phase unwrapping; region growing; synthetic aperture radar (SAR); terrain mapping;
D O I
10.1109/36.739143
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
This paper describes a new region-growing algorithm for interferometric synthetic aperture radar (SAR) phase unwrapping. The algorithm is designed to handle noisy interferograms and based on the following principles. 1) Unwrapping is carried out on the perimeter of "growth regions," and these regions are allowed to grow with consistency checking. 2) Phase information from neighboring pixels is used to predict the correct phase of each new pixel to be un wrapped. 3) Reliability check is applied to each unwrapping attempt, and the threshold used in the cheek is gradually relaxed so that the most robust unwrapping path is followed, 4) As regions grow into one another, they are merged by adjusting their ambiguity numbers, and further reliability cheeks must be passed for merging to be allowed. To verify the proposed algorithm, the region-growing phase unwrapping algorithm has been compared with two of the most widely used algorithms: the cut-line (CL) (also known as residue-linking) algorithm and the weighted least-squares (WLS) algorithm. The FRINGE Sardegna interferogram has been used because it has steep topography that causes problems for many phase unwrapping algorithms, It has been shown that the region-growing algorithm can unwrap well into low coherence regions, while making few ambiguity-level errors. This can be attributed to the diversity of growth paths used by the algorithm, in which difficult areas are approached from many directions.
引用
收藏
页码:124 / 134
页数:11
相关论文
共 22 条
[1]   Noise-induced slope distortion in 2-D phase unwrapping by linear estimators with application to SAR interferometry [J].
Bamler, R ;
Adam, N ;
Davidson, GW ;
Just, D .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 1998, 36 (03) :913-921
[2]  
COSTANTINI M, P FRINGE 9L ESA WORK
[3]  
CUMMINGS I, 1997, P 3 ERS SCI S FLOR I
[4]  
DAVIDSON GW, P IGARSS 96 LINC NE, P2050
[5]  
*ERS ESRIN, 1992, P ERS 1 FRINGE 1 WOR
[6]  
FLYNN TJ, P IGARSS 9L LINC NE, P2057
[7]  
Ghiglia D. C., 1996, J OPT SOC AM A, V13, P1
[8]   ROBUST 2-DIMENSIONAL WEIGHTED AND UNWEIGHTED PHASE UNWRAPPING THAT USES FAST TRANSFORMS AND ITERATIVE METHODS [J].
GHIGLIA, DC ;
ROMERO, LA .
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION, 1994, 11 (01) :107-117
[9]   SATELLITE RADAR INTERFEROMETRY - TWO-DIMENSIONAL PHASE UNWRAPPING [J].
GOLDSTEIN, RM ;
ZEBKER, HA ;
WERNER, CL .
RADIO SCIENCE, 1988, 23 (04) :713-720
[10]  
JOUGHIN IR, 1995, THESIS U WASHINGTON