Polarimetric Feature Comparison of Full and Dual Polarimetric SAR

Document Type : Original Article

Authors

1 Assistant prof. of Remote Sensing and GIS Dep., Faculty of Geography, University of Tehran

2 Assistant prof. of Atmospheric Science and Meteorological Research Center

Abstract

Full polarimetric SAR sensors can capture full polarimetric characteristics of targets. Therefore, in comparison with single and dual polarimetric sensors they offer more capabilities in target detection. However, operation in full polarimetric mode increases complexity, data volume and need more power. Full polarimetric sensors acquire images with less swath compared to dual mode. As a result, most of SAR sensors operate in dual mode and provide dual polarimetric images. Due to high availability, dual polarimetric images are increasingly being used in many researches. In this research, the efficiency of dual polarimetric images is compared with full polarimetric mode. The main goal is to find the best combination of two polarimetric bands which has the nearest results to full polarimetric mode.One Advanced Land Observing Satellite / Phased Array L-band Synthetic Aperture Radar scene had been processed. The scene was multi-looked and converted to the backscattering coefficient (sigma nought, dB). The image was decomposed by cluode-pottier method into alpha and entropy components. Three different combination of two polarimetric bands were considered; HH-HV; HH-VV and HV-VV. Alpha and entropy of each dual polarimetric mode were also computed. Then alpha and entropy driven from full-polarimetric mode were separately compared with alpha and entropy of each dual mode. Since different land cover types (i.e. built-up, cropland, bare land and water) exist in the scene, the computations were done separately for each land cover type. The comparison among alpha values from full polarimetric mode and dual polarimetric mode reveals that HH-HV combination shows the best conformity with full polarimetric mode. HH-VV dual mode has the poorest results. Entropy values of HH-HV mode had the least difference with full polarimetric mode. Entropy values of HH-VV shows the weakest similarity. The MAE values of HH-HV, HH-VV and HV-VV were 0.06, 0.22 and 0.17, respectively. The findings of this research shows that polarimetric features driven from HH-HV combination are more compatible with full-polarimetric mode. In case, no full polarimetric image is available, this dual combination can be substituted. Based on quantitative results, HH-HV combination is recommended to be used in case no full polarimetric image is availableOne Advanced Land Observing Satellite / Phased Array L-band Synthetic Aperture Radar scene had been processed. The scene was multi-looked and converted to the backscattering coefficient (sigma nought, dB). The image was decomposed by cluode-pottier method into alpha and entropy components. Three different combination of two polarimetric bands were considered; HH-HV; HH-VV and HV-VV. Alpha and entropy of each dual polarimetric mode were also computed. Then alpha and entropy driven from full-polarimetric mode were separately compared with alpha and entropy of each dual mode. Since different land cover types (i.e. built-up, cropland, bare land and water) exist in the scene, the computations were done separately for each land cover type. The comparison among alpha values from full polarimetric mode and dual polarimetric mode reveals that HH-HV combination shows the best conformity with full polarimetric mode. HH-VV dual mode has the poorest results. Entropy values of HH-HV mode had the least difference with full polarimetric mode. Entropy values of HH-VV shows the weakest similarity. The MAE values of HH-HV, HH-VV and HV-VV were 0.06, 0.22 and 0.17, respectively. The findings of this research shows that polarimetric features driven from HH-HV combination are more compatible with full-polarimetric mode. In case, no full polarimetric image is available, this dual combination can be substituted. Based on quantitative results, HH-HV combination is recommended to be used in case no full polarimetric image is available.

Keywords


Ainsworth, T.L., Kelly, J.P. & Lee, J.S., 2009, Classification Comparisons between Dual-pol, Compact Polarimetric and Quad-pol SAR Imagery, ISPRS Journal of Photogrammetry and Remote Sensing, 64(5), PP. 464-471.
Cloude, S., 2007, The Dual Polarization Entropy/Alpha Decomposition: A PALSAR Case Study, Science and Applications of SAR Polarimetry and Polarimetric Interferometry,  Vol. 644.
Cloude, S.R. & Pottier, E., 1996, A revIew of Target Decomposition Theorems in Radar Polarimetry, IEEE Transactions on Geoscience and Remote Sensing, 34(2), PP. 498-518.
Dhar, A.T., Gray, B.D. & Menges, C.C., 2011, Comparison of Dual and Full Polarimetric Entropy/Alpha Decompositions with TerraSAR-X, Suitability for Use in Classification, In 2011 IEEE International Geoscience and Remote Sensing Symposium,  PP. 456-458, IEEE.
Freeman, A. & Durden, S.L., 1998, A Three-component Scattering Model for Polarimetric SAR Data, IEEE Transactions on Geoscience and Remote Sensing, 36(3), PP. 963-973.
Izumi, Y., Demirci, S., bin Baharuddin, M., Watanabe, T. & Sumantyo, J., 2017, Analysis of Dual-and Full-circular Polarimetric SAR Modes for Rice Phenology Monitoring: An Experi-mental Investigation through Ground-based Measurements, Applied Sciences, 7(4), P. 368.
Ji, K. & Wu, Y., 2015, Scattering Mechanism Extraction by a Modified Cloude-pottier Decomposition for Dual Polarization SAR, Remote Sensing, 7(6), PP. 7447-7470.
Lavalle, M. & Wright, T., 2009, Absolute Radiometric and Polarimetric Calibra-tionofALOS PALSAR Products, Document lssue (1), Revision (3).
Lee, J.S., Grunes, M.R. & Pottier, E., 2001, Quantitative Comparison of Classifica-tion Capability: Fully Polarimetric Versus Dual and Single-polarization SAR, IEEE Transactions on Geoscience and Remote Sensing, 39(11), PP. 2343-2351.
Lee, J.S., Grunes, M.R., Pottier, E. & Ferro-Famil, L., 2004, Unsupervised Terrain Classification Preserving Polarimetric Scattering Characteristics, IEEE Tran-sactions on Geoscience and Remote Sensing, 42(4), PP. 722-731.
 
Łoś, H., Osińska-Skotak, K., Pluto-Kossakowska, J., Bernier, M., Gauthier, Y. & Pawłowski, B., 2019, Performance Evaluation of Quad-pol Data Compare to Dual-pol SAR Data for River Ice Classification, European Journal of Remote Sensing, 52(sup1), PP. 79-95.
Shan, Z., Wang, C., Zhang, H. & Chen, J., 2011, H-alpha Decomposition and Alternative Parameters for Dual Polarization SAR Data, Proc. PIERS, SuZhou, China.
Simioni, J.P.D., Guasselli, L.A., Nascimento, V.F., Ruiz, L.F.C. & Belloli, T.F., 2019, Integration of Multi-sensor Analysis and Decision Tree for Evaluation of Dual and Quad-Pol SAR in L-and C-bands Applied for Marsh Delineation, Environment, Development and Sustaina-bility, 1-18.
Sugimoto, M., Ouchi, K. & Nakamura, Y., 2013, On the Similarity between Dual-and Quad-eigenvalue Analysis in SAR Polarimetry, Remote Sensing Letters, 4(10), PP. 956-964.
Touzi, R., Goze, S., Le Toan, T., Lopes, A. & Mougin, E.,1992, Polarimetric Discrimi-nators for SAR Images, IEEE Transactions on Geoscience and Remote Sensing, 30(5), PP. 973-980.
Turkar, V., Deo, R., Hariharan, S. & Rao, Y.S., 2011, Comparison of Classification Accuracy between Fully Polarimetric and Dual-polarization SAR Images, In 2011 IEEE International Geoscience and Remote Sensing Symposium, PP. 440-443.
Van Zyl, J.J., 1989, Unsupervised Classification of Scattering Behavior Using Radar Polarimetry Data, IEEE Transactions on Geoscience and Remote Sensing, 27(1), PP. 36-45.
Xie, L., Zhang, H., Wang, C. & Shan, Z., 2015, Similarity Analysis of Entropy/ alpha Decomposition between HH/VV Dual-and Quad-polarization SAR Data, Remote Sensing Letters, 6(3), PP. 228-237.
Yamaguchi, Y., Moriyama, T., Ishido, M. & Yamada, H., 2005, Four-component Scattering Model for Polarimetric SAR Image Decomposition, IEEE Transactions on Geoscience and Remote Sensing, 43(8), PP. 1699-1706.