Unsupervised classification using polarimetric decomposition and the complex Wishart classifier

Lee, Jong-Sen; Grunes, M.R.; Ainsworth, T.L.; Du, Li-Jen; Schuler, D.L.; Cloude, S.R.

doi:10.1109/36.789621

Cited by 756 publications

(66 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As can be observed in Figure 8, the extensive sand spit was not discriminated from adjacent water at steep angles, but can be discerned at shallow and medium angles. This is consistent with observations made by Lee et al (1999) using L-band AIRSAR data acquired over San Francisco, California. Shelat et al (2012) also observed similar results using an FQ1 RADARSAT-2 scene to classify Arctic surficial materials.…”

Section: Wishart-entropy/alpha Wishart-entropy/anisotropy/alpha Classupporting

confidence: 91%

“…A well-known limitation of the entropy-alpha segmentation is that the physical boundaries proposed by Cloude and Pottier (1997) do not always coincide with natural breaks in the data (Lee et al 1999). As described by Cloude and Pottier (1997), selection of these boundaries was somewhat arbitrary, and at the time they were proposed it was also anticipated that a number of data and sensor parameters, such as measurement noise floor, would have an impact on their definition.…”

Section: Cloude-pottier Decompositionmentioning

confidence: 98%

“…This classification also represents the first step of the W H,α classifier, which uses the clusters that are produced to train a subsequent Wishart classifier (Wishart distance-based maximum likelihood), applied in multiple iterations. This classifier may assign pixels to new clusters, making it difficult to use results for interpreting scattering behaviour (Lee et al 1999;Lee and Pottier 2009). The W H,A,α classifier works similarly, but anisotropy values are used to generate an additional 8 clusters during the initial classification.…”

Section: Polarimetric Decompositionsmentioning

confidence: 99%

See 2 more Smart Citations

Characterizing Scattering Behaviour and Assessing Potential for Classification of Arctic Shore and Near-Shore Land Covers with Fine Quad-Pol RADARSAT-2 Data

Banks

King

Merzouki

et al. 2014

Canadian Journal of Remote Sensing

View full text Add to dashboard Cite

Freeman-Durden and Cloude-Pottier decompositions were applied to polarimetric RADARSAT-2 data to characterize the scattering behaviour of shore and near-shore land cover types for two study areas in the Beaufort Sea, Canada: Richards Island and Tuktoyaktuk Harbour. The impact of incidence angle was evaluated through comparison of Single Look Complex Fine Quad-Pol images acquired at three angles: shallow (45.3 • -49.5 • ), medium (39.3 • -41.6 • ), and steep (20.9 • -24.2 • ), and the potential for accurate classification was assessed using three unsupervised polarimetric SAR classifiers: Wishart-entropy/alpha, Wishart-entropy/anisotropy/alpha, and Freeman-Wishart. Results indicate that overall, medium, and shallow incidence angle images provide more useful information than steep angle images and that these classifiers are capable of distinguishing some general land cover types, including substrates, wetlands, vegetated tundra, and water. Results from this analysis will inform future research related to classification and interpretation of RADARSAT-2 data for shoreline mapping. Résumé. Des décompositions Freeman-Durden etCloude-Pottier ontété appliquéesà des données polarimétriques RADARSAT-2 pour caractériser le comportement de diffusion des différents types de couverture terrestre des milieux littoraux et côtiers pour deux zones d'étude dans la mer de Beaufort, au Canada: l'île Richards et le port de Tuktoyaktuk. L'impact de l'angle d'incidence aétéévalué par comparaison des images singulières complexes <> quad-pol fines acquisesà trois angles; plus rasants (45.3 • -49.5 • ), moyens (39.3 • -41.6 • ) et abrupts (20.9 • -24.2 • ), tandis que le potentiel pour une classification précise aétéévaluéà l'aide de trois classificateurs RSO polarimétriques non dirigés: Wishart-entropie/alpha, Wishart-entropie/anisotropie/alpha et Freeman-Wishart.Les résultats indiquent que, dans l'ensemble, les images avec des angles d'incidence moyensà rasants fournissent des informations plus utiles que les images aux angles abrupts, et que ces classificateurs sont capables de distinguer certains types généraux de couverture terrestre, y compris: les substrats, les zones humides, la toundra végétalisée et l'eau. Les résultats de cette analyse seront utiles dans les recherches futures liéesà la classification età l'interprétation des données RADARSAT-2 pour la cartographie des rivages.

show abstract

Section: Wishart-entropy/alpha Wishart-entropy/anisotropy/alpha Classupporting

confidence: 91%

Section: Cloude-pottier Decompositionmentioning

confidence: 98%

Section: Polarimetric Decompositionsmentioning

confidence: 99%

See 1 more Smart Citation

Characterizing Scattering Behaviour and Assessing Potential for Classification of Arctic Shore and Near-Shore Land Covers with Fine Quad-Pol RADARSAT-2 Data

Banks

King

Merzouki

et al. 2014

Canadian Journal of Remote Sensing

View full text Add to dashboard Cite

show abstract

“…In the pixel-based methods, features are derived on a pixel by pixel basis, supervised and unsupervised classification schemes generally use Wishart statistics to provide a *Corresponding author. Email: lulj@casm.ac.cn International Journal of Image and Data Fusion, 2015http://dx.doi.org/10.1080/19479832.2015 maximum likelihood decision rule on the class occupancy of image pixels (Lee et al 1999. Unsupervised classification employs some polarimetric decomposition parameters, for example, H, alpha and A.…”

Section: Introductionmentioning

confidence: 99%

Land cover classification and height extraction experiments using Chinese airborne X-band PolInSAR system in China

Zhang

Huang

et al. 2015

International Journal of Image and Data Fusion

View full text Add to dashboard Cite

In this paper, we describe our land cover classification and height extraction experiments with high-resolution X-band polarimetric interferometric synthetic aperture radar (PolInSAR) data acquired by Chinese airborne PolInSAR system over Linshui in southern China. The experimental area contains plenty of natural and artificial objects. We present two polarimetric synthetic aperture radar (PolSAR) classification schemes for different land cover types. An Oriented-Object classification scheme based on hierarchical segmentation is employed to the regular land cover classification experiment. The Entropy-PSD unsupervised classification method based on pixel is employed to the low backscattering targets classification experiment. Fully taking advantage of the characteristics of the multi-baseline data set, a dual-baseline polarimetric interferometry method is applied to the PolInSAR experiment of height extraction. Our PolSAR and PolInSAR experiment results demonstrate that high-resolution X-band PolInSAR imagery acquired by Chinese airborne SAR system have great application potential for land classification and height extraction.

show abstract

“…The use of FP data in land-cover classification has been widely assessed using both airborne and spaceborne data. Lee et al (1999Lee et al ( , 2004 and Ferro-Famil, Pottier, and Lee (2001) used NASA/Jet Propulsion Laboratory (JPL) Airborne SAR (AirSAR) data for crop classification. Tu et al (2012) explored X-band airborne data for classification.…”

Section: Introductionmentioning

confidence: 99%

A unified framework for crop classification in southern China using fully polarimetric, dual polarimetric, and compact polarimetric SAR data

Xie

Zhang

et al. 2015

International Journal of Remote Sensing

View full text Add to dashboard Cite

With the development of synthetic aperture radar (SAR) techniques, various imaging modes that involve single polarimetry, dual polarimetry, full polarimetry (FP), and compact polarimetry (CP) have been proposed and applied to SAR systems. This article attempts to introduce a unified framework for crop classification in southern China using FP, coherent HH/VV, and CP data. By analysing the polarimetric response from different land-cover types (including rice, banana trees, sugarcane, eucalyptus, water, and built-up areas in the experimental site) and by exploring the similarities between data in these three modes, a knowledge-based characteristic space is created and a unified classification framework is presented. Time-series data acquired by TerraSAR-X over the Leizhou Peninsula, southern China, are used in our experiments. The overall classification accuracies for data in the FP and coherent HH/VV modes are approximately 95%, and for data in the CP mode, the accuracy is 91%, which suggest that the proposed classification scheme is effective. Compared with the Wishart Maximum Likelihood (ML) classifier, the proposed method provides approximately 5.64%, 7.30%, and 6.48% higher classification accuracies in the FP, HH/VV, and circular transmit and dual circular receive modes, respectively.

show abstract

Unsupervised classification using polarimetric decomposition and the complex Wishart classifier

Cited by 756 publications

References 19 publications

Characterizing Scattering Behaviour and Assessing Potential for Classification of Arctic Shore and Near-Shore Land Covers with Fine Quad-Pol RADARSAT-2 Data

Characterizing Scattering Behaviour and Assessing Potential for Classification of Arctic Shore and Near-Shore Land Covers with Fine Quad-Pol RADARSAT-2 Data

Land cover classification and height extraction experiments using Chinese airborne X-band PolInSAR system in China

A unified framework for crop classification in southern China using fully polarimetric, dual polarimetric, and compact polarimetric SAR data

Contact Info

Product

Resources

About