Forest classification using extracted PolSAR features from Compact Polarimetry data

Abstract This study investigates the ability of extracted Polarimetric Synthetic Aperture RADAR (PolSAR) features from Compact Polarimetry (CP) data for forest classification. The CP is a new mode that is recently proposed in Dual Polarimetry (DP) imaging system. It has several important advantages in comparison with Full Polarimetry (FP) mode such as reduction ability in complexity, cost, mass, data rate of a SAR system. Two strategies are employed for PolSAR feature extraction. In first strategy, the features are extracted using 2 × 2 covariance matrices of CP modes simulated by RADARSAT-2 C-band FP mode. In second strategy, they are extracted using 3 × 3 covariance matrices reconstructed from the CP modes called Pseudo Quad (PQ) modes. In each strategy, the extracted PolSAR features are combined and optimal features are selected by Genetic Algorithm (GA) and then a Support Vector Machine (SVM) classifier is applied. Finally, the results are compared with the FP mode. Results of this study show that the PolSAR features extracted from π / 4 CP mode, as well as combining the PolSAR features extracted from CP or PQ modes provide a better overall accuracy in classification of forest.

[1]  Shaun Quegan,et al.  Faraday rotation effects on L-band spaceborne SAR data , 2003, IEEE Trans. Geosci. Remote. Sens..

[2]  Hao Chen,et al.  Compact Decomposition Theory , 2012, IEEE Geoscience and Remote Sensing Letters.

[3]  Sergios Theodoridis,et al.  Chapter 11 – Clustering: Basic Concepts , 2006 .

[4]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[5]  Christopher J. C. Burges,et al.  A Tutorial on Support Vector Machines for Pattern Recognition , 1998, Data Mining and Knowledge Discovery.

[6]  A. Belhadj-aissa,et al.  Investigation of the capability of the Compact Polarimetry mode to Reconstruct Full Polarimetry mode using RADARSAT2 data , 2012 .

[7]  Jixian Zhang,et al.  Land Cover Classification from Polarimetric SAR Data Based on Image Segmentation and Decision Trees , 2015 .

[8]  Manab Chakraborty,et al.  ASSESSMENT OF L-BAND SAR DATA AT DIFFERENT POLARIZATION COMBINATIONS FOR CROP AND OTHER LANDUSE CLASSIFICATION , 2012 .

[9]  Yasser Maghsoudi,et al.  Polarimetric classification of Boreal forest using nonparametric feature selection and multiple classifiers , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[10]  Jean-Claude Souyris,et al.  Compact polarimetry based on symmetry properties of geophysical media: the /spl pi//4 mode , 2005, IEEE Transactions on Geoscience and Remote Sensing.

[11]  Mohammed Dabboor,et al.  Towards sea ice classification using simulated RADARSAT Constellation Mission compact polarimetric SAR imagery , 2014 .

[12]  Sébastien Angélliaume,et al.  The Compact Polarimetry Alternative for Spaceborne SAR at Low Frequency , 2008, IEEE Transactions on Geoscience and Remote Sensing.

[13]  D. H. Hoekman,et al.  Interpretation of C- and X-band radar images over an agricultural area, the Flevoland test site in the Agriscatt-87 campaign. , 1993 .

[14]  Torsten Geldsetzer,et al.  Compact Polarimetry in Support of Lake Ice Breakup Monitoring: Anticipating the RADARSAT Constellation Mission , 2015 .

[15]  Bambang H. Trisasongko.  Evaluating compact SAR polarimetry for tropical forest monitoring , 2015, International Seminar on Photonics, Optics, and its Applications.

[16]  Dong Sun,et al.  Combined Similarity-Based Spectral Clustering Ensemble for PolSAR Land Cover Classification , 2015 .

[17]  A. Mansourian,et al.  Polarimetric SAR feature selection using a genetic algorithm , 2011 .

[18]  Lalit Mohan Saini,et al.  A critical analysis of EM based fusion of different polarization data for effect on land cover classification , 2015 .

[19]  Jaan Praks,et al.  Land Cover and Soil Type Mapping From Spaceborne PolSAR Data at L-Band With Probabilistic Neural Network , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[20]  Jong-Sen Lee,et al.  Classification comparisons between dual-pol, compact polarimetric and quad-pol SAR imagery , 2009 .

[21]  A. O. Varghese,et al.  Polarimetric Classification of C-Band SAR Data for forest Density Characterization , 2015 .

[22]  E. Pottier,et al.  Polarimetric Radar Imaging: From Basics to Applications , 2009 .

[23]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[24]  Christian Thiel,et al.  Mapping CORINE Land Cover from Sentinel-1A SAR and SRTM Digital Elevation Model Data using Random Forests , 2015, Remote. Sens..

[25]  S. Boukir,et al.  Texture-based forest cover classification using random forests and ensemble margin , 2015, 2015 IEEE International Geoscience and Remote Sensing Symposium (IGARSS).

[26]  Sergios Theodoridis,et al.  Clustering: Basic Concepts , 2009 .

[27]  Bernhard Schölkopf,et al.  Feature selection for support vector machines by means of genetic algorithm , 2003, Proceedings. 15th IEEE International Conference on Tools with Artificial Intelligence.

[28]  Paul G. Jarvis,et al.  12 – Forests in the Global Carbon Balance: From Stand to Region , 1993 .

[29]  Heather McNairn,et al.  Compact polarimetry overview and applications assessment , 2010 .

[30]  R. Keith Raney,et al.  The m-chi decomposition of hybrid dual-polarimetric radar data , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[31]  Fuk K. Li,et al.  Symmetry properties in polarimetric remote sensing , 1992 .

[32]  Mohammed Dabboor,et al.  Change Detection with Compact Polarimetric SAR for Monitoring Wetlands , 2015 .