Unsupervised Synthetic Aperture Radar Image Segmentation Using Fisher Distributions

被引：47

作者：

Galland, Frederic ^{[1
]}

Nicolas, Jean-Marie ^{[2
]}

Sportouche, Helene ^{[1
,2
]}

Roche, Muriel ^{[1
]}

Tupin, Florence ^{[2
]}

Refregier, Philippe ^{[1
]}

机构：

[1] Aix Marseille Univ, Ecole Cent Marseille, CNRS, Inst Fresnel, F-13013 Marseille, France

[2] CNRS LTCI, TELECOM ParisTech, Inst TELECOM, F-75634 Paris 13, France

来源：

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING | 2009年 / 47卷 / 08期

关键词：

Fisher distribution; minimum description length; segmentation; stochastic complexity; synthetic aperture radar (SAR); SAR; MODEL; CLASSIFICATION; ESTIMATORS;

D O I：

10.1109/TGRS.2009.2014364

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：

A new and fast unsupervised technique for segmentation of high-resolution synthetic aperture radar (SAR) images into homogeneous regions is proposed. This technique is based on Fisher probability density functions (pdfs) of the intensity fluctuations and on an image model that consists of a patchwork of homogeneous regions with polygonal boundaries. The segmentation is obtained by minimizing the stochastic complexity of the image. Different strategies for the pdf parameter estimation are analyzed, and a fast and robust technique is proposed. Finally, the relevance of the proposed approach is demonstrated on high-resolution SAR images.

引用

页码：2966 / 2972

页数：7

共 29 条

[1]

[Anonymous], 1989, STOCHASTIC COMPLEXIT

[2]

[Anonymous], UNDERSTANDING SYNTHE

[3] Multiregion level-set partitioning of synthetic aperture radar images [J].

Ben Ayed, I ;

Mitiche, A ;

Belhadj, Z .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2005, 27 (05) :793-800

[4] Parameter estimation for the K-distribution based on [z log(z)] [J].

Blacknell, D ;

Tough, RJA .

IEE PROCEEDINGS-RADAR SONAR AND NAVIGATION, 2001, 148 (06) :309-312

[5] Fisher distribution for texture modeling of polarimetric SAR data [J].

Bombrun, Lionel ;

Beaulieu, Jean-Marie .

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2008, 5 (03) :512-516

[6] Statistical region snake-based segmentation adapted to different physical noise models [J].

Chesnaud, C ;

Réfrégier, P ;

Boulet, V .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1999, 21 (11) :1145-1157

[7] On the design and evaluation of multiobjective single-channel SAR image segmentation algorithms [J].

Collins, Michael J. ;

Kopp, Eric B. .

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2008, 46 (06) :1836-1846

[8]

COOK R, 1994, P SOC PHOTO-OPT INS, V2316, P92, DOI 10.1117/12.197529

[9]

Cover Thomas M., 2006, Elements of Information Theory, V2nd

[10]

DUBOISFERNANDEZ P, 2002, P IGARSS TOR ON CAN, P1723

← 1 2 3 →