Classification using active polarimetry

Active (Mueller matrix) remote sensing is an under-utilized technique for material discrimination and classication. A full Mueller matrix instrument returns more information than a passive (Stokes) polarimeter; Mueller polarimeters measure depolarization and other linear transformations that materials impart on incident Stokes vectors, which passive polarimeters cannot measure. This increase in information therefore allows for better classication of materials (in general). Ideally, material classication over the entire polarized BRDF is desired, but sets of Mueller matrices for dierent materials are generally not separable by a linear classier over elevation and azimuthal target angles. We apply non-linear support vector machines (SVM) to classify materials over BRDF (all relevant angles) and show variations in receiver operator characteristic curves with scene composition and number of Mueller matrix channels in the observation.

[1]  J. Zyl,et al.  Unsupervised classification of scattering behavior using radar polarimetry data , 1989 .

[2]  Dennis H. Goldstein,et al.  Reflective and polarimetric characteristics of urban materials , 2006, SPIE Defense + Commercial Sensing.

[3]  Yoshitate Takakura,et al.  Noise distribution of Mueller matrices retrieved with active rotating polarimeters. , 2007, Applied optics.

[4]  François Goudail,et al.  Comparison of maximal achievable contrast in scalar, Stokes, and Mueller images. , 2010, Optics letters.

[5]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[6]  Kurt Keutzer,et al.  Fast support vector machine training and classification on graphics processors , 2008, ICML '08.

[7]  Ron Kwok,et al.  Classification of multi-look polarimetric SAR imagery based on complex Wishart distribution , 1994 .

[8]  Hongbin Zha,et al.  Riemannian Manifold Learning , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  G. Gustafson,et al.  Boundary Value Problems of Mathematical Physics , 1998 .

[10]  J Scott Tyo,et al.  Polarization components analysis for invariant discrimination. , 2007, Applied optics.

[11]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[12]  Laurent Ferro-Famil,et al.  Unsupervised classification of multifrequency and fully polarimetric SAR images based on the H/A/Alpha-Wishart classifier , 2001, IEEE Trans. Geosci. Remote. Sens..

[13]  L. Zangwill,et al.  Detection of glaucoma with scanning laser polarimetry. , 1998, Archives of ophthalmology.

[14]  François Goudail,et al.  Optimal discrimination of multiple regions with an active polarimetric imager. , 2011, Optics express.

[15]  Eric Pottier,et al.  An entropy based classification scheme for land applications of polarimetric SAR , 1997, IEEE Trans. Geosci. Remote. Sens..

[16]  Russell A. Chipman,et al.  Error analysis of a Mueller matrix polarimeter , 1990 .

[17]  Kamal Sarabandi,et al.  Knowledge-based classification of polarimetric SAR images , 1994, IEEE Trans. Geosci. Remote. Sens..

[18]  Mark A. Hall,et al.  Correlation-based Feature Selection for Discrete and Numeric Class Machine Learning , 1999, ICML.

[19]  R. Chipman,et al.  Optimization of Mueller matrix polarimeters in the presence of error sources. , 2008, Optics express.

[20]  A. Nashashibi,et al.  Statistical properties of the Mueller matrix of distributed targets. [measured in backscattering direction by polarimetric radar system] , 1992 .

[21]  Matthew H. Smith,et al.  Optimization of a dual-rotating-retarder Mueller matrix polarimeter. , 2002, Applied optics.

[22]  Israel J. Vaughn,et al.  Noise reduction in a laser polarimeter based on discrete waveplate rotations. , 2008, Optics express.

[23]  F. Goudail,et al.  When is polarimetric imaging preferable to intensity imaging for target detection? , 2011, Journal of the Optical Society of America. A, Optics, image science, and vision.

[24]  François Goudail,et al.  Polarimetric target detection in the presence of spatially fluctuating Mueller matrices. , 2011, Optics letters.