“science value represents the potential of the collected data to be turned into information of geophysical significance.”– Dr. R. K. Raney, Life Fellow, IEEE
Target Characterization and Scattering Power Decomposition for Full and Compact Polarimetric SAR Data
In radar polarimetry, incoherent target decomposition techniques help extract scattering information from polarimetric synthetic aperture radar (SAR) data. This is achieved either by fitting appropriate scattering models or by optimizing the received wave intensity through the diagonalization of the coherency (or covariance) matrix. As such, the received wave information depends on the received antenna configuration. Additionally, a polarimetric descriptor that is independent of the received antenna configuration might provide additional information which is missed by the individual elements of the coherency matrix. This implies that existing target characterization techniques might neglect this information. In this regard, we suitably utilize the 2-D and 3-D Barakat degree of polarization which is independent of the received antenna configuration to obtain distinct polarimetric information for target characterization. In this study, we introduce new roll-invariant scattering-type parameters for both full-polarimetric (FP) and compact-polarimetric (CP) SAR data. These new parameters jointly use the information of the 2-D and 3-D Barakat degree of polarization and the elements of the coherency (or covariance) matrix. We use these new scattering-type parameters, which provide equivalent information as the Cloude α for FP SAR data and the ellipticity parameter χ for CP SAR data, to characterize various targets adequately. Additionally, we appropriately utilize these new scattering-type parameters to obtain unique non-model-based three-component scattering power decomposition techniques. We obtain the even-bounce, and the odd-bounce scattering powers by modulating the total polarized power by a proper geometrical factor derived using the new scattering-type parameters for FP and CP SAR data. The diffused scattering power is obtained as the depolarized fraction of the total power. Moreover, due to the nature of its formulation, the decomposition scattering powers are non-negative and roll-invariant while the total power is conserved. The proposed method is both qualitatively and quantitatively assessed utilizing the L-band ALOS-2 and C-band Radarsat-2 FP and the associated simulated CP SAR data.Dey, S., Bhattacharya, A., Ratha, D., Mandal, D. and Frery, A.C., 2020. Target Characterization and Scattering Power Decomposition for Full and Compact Polarimetric SAR Data. IEEE Transactions on Geoscience and Remote Sensing.
Novel Techniques for Built-Up Area Extraction From Polarimetric SAR Images
Target descriptions provided by full-polarimetric SAR (PolSAR) data might be even more useful. Nevertheless, they are hard to employ due to the complex interactions in urban areas among the electromagnetic waves, the targets, and their surroundings. Therefore, the main avenues for extracting built-up area information from PolSAR imagery range from using decomposition theorems to feeding the data into deep learning architectures.
From the processing point of view, these state-of-the-art methods span from decision trees to mixture models to deep learning, respectively. In general, they require either long processing steps for backscatter analysis or manual setting of the hyper-parameters. This letter aims at completing this scenario with a technique that is direct and intuitive, as well as easy to implement. Another objective is introducing a novel pixel-based BU index in PolSAR based on exploiting scattering mechanisms.
Fig: Built-up area maps for RADARSAT-2 C-band image over San Francisco, USA. (left) BU index. (right) BU area map.
We propose a scattering similarity-based approach using the observed data and specific elementary scattering models. Structures within built-up areas show a high degree of similarity with narrow dihedral and dihedral models, while the cross-polarized component exists only with the helix models. In addition, a new radar built-up index (RBUI) is proposed and validated.D. Ratha, P. Gamba, A. Bhattacharya and A. C. Frery, “Novel Techniques for Built-Up Area Extraction From Polarimetric SAR Images,” in IEEE Geoscience and Remote Sensing Letters. DOI: 10.1109/LGRS.2019.2914913
Snow Cover Mapping Using Polarization Fraction Variation
Snow is a dynamic matter since its dielectric state is dependent on climatic factors prevailing locally around it. In the literature, the polarization fraction (PF) has been used for landcover characterization. In this paper, we utilize the seasonal variation of the PF for mapping snow cover over the Himalayan terrain. Temporal variation of the strength in the polarized return due to change in landcover and season is the prime motivation behind this approach. In addition, the effect of SAR data acquisition time on mapping algorithms is considered in this work which is seldom discussed in the literature.
Fig: Snow cover maps derived from RADARSAT-2. (a) 22nd February 2015. (b) 18th March 2015.
The applicability of the approach is tested with RADARSAT-2 (FQ28) C-band full polarimetric image pairs over the Manali–Dhundi region, Himachal Pradesh, India, located in the western Hindu-Kush Himalayas.A. Muhuri, S. Manickam, A. Bhattacharya and Snehmani, “Snow Cover Mapping Using Polarization Fraction Variation With Temporal RADARSAT-2 C-Band Full-Polarimetric SAR Data Over the Indian Himalayas,” in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 11, no. 7, pp. 2192-2209, July 2018. DOI: 10.1109/JSTARS.2018.2817687
Relative Decorrelation Measure in PolSAR Decomposition (RD-Y4O)
A novel methodology is proposed to improve the scattering powers obtained from model-based decomposition using Polarimetric Synthetic Aperture Radar (PolSAR) data. The novelty of this approach lies in utilizing the intrinsic information in the off-diagonal elements of the 3 × 3 coherency matrix T represented in the form of complex correlation coefficients. Two complex correlation coefficients are computed between co-polarization and cross-polarization components of the Pauli scattering vector. The difference between modulus of complex correlation coefficients corresponding to T^opt (i.e., the degree of polarization (DOP) optimized coherency matrix), and T (original) matrices is obtained. Then a suitable scaling is performed using fractions obtained from the diagonal elements of the T^opt matrix. Thereafter, these new quantities are used in modifying the Yamaguchi 4-component scattering powers obtained from T^opt.
Fig: ALOS PALSAR-2 L-band Kyoto image: (a) Y4O (b) Y4R (c) RD–Y4O; (d), (e) and (f) are corresponding scattering powers for the red rectangle.
To corroborate the fact that these quantities have physical relevance, a quantitative analysis of these for the L-band AIRSAR San Francisco and the L-band Kyoto images is illustrated. Finally, the scattering powers obtained from the proposed methodology are compared with the corresponding powers obtained from the Yamaguchi et. al., 4-component decomposition and the Yamaguchi et. al., 4-component Rotated decomposition for the same data sets. The proportion of negative power pixels is also computed. The results show an improvement on all these attributes by using the proposed methodology.D. Ratha, M. Surendar & A. Bhattacharya (2017) Improvement of PolSAR decomposition scattering powers using a relative decorrelation measure, Remote Sensing Letters, 8:4, 340-349, DOI: 10.1080/2150704X.2016.1271159
Adaptive Generalized Four Component Decomposition with Unitary Transformation (AG4U)
An adaptive general four-component scattering power decomposition method (AG4U) is proposed in this letter. The degree of polarization mis used as a criterion for the adaptive nature of the proposed decomposition. In this method, one among the two complex special unitary transformation matrices is chosen to transform a real unitary rotated coherency matrix based on the largest value of m. This transformed matrix is then utilized for the existing Yamaguchi et al. four-component decomposition scheme with an extended volume scattering model.
Fig: UAVSAR L-band Hayward image. (a) Yamaguchi four-component decomposition without rotation of the coherency matrix (Y4O). (b) General four-component scattering power decomposition with unitary transformation (G4U). (c) Adaptive general four-component scattering power decomposition with unitary transformation (AG4U). The scattering powers for areas “A” and “B” are shown in (d) and (e), respectively.
The proposed decomposition is applied to Radarsat-2 full-polarimetic C-band data over San Francisco and Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) full-polarimetric L-band data over the Hayward Fault in California. The scattering powers estimated from the decomposition techniques of Yamaguchi et al. (Y4O), Singh et al. (G4U), and AG4U are compared. AG4U shows appreciable improvements in the scattering powers, particularly in urban areas oriented about the radar line of sight compared with the Y4O and G4U decompositions. It also shows reduced percentage of pixels with negative powers considerably compared with the Y4O decomposition.A. Bhattacharya, G. Singh, S. Manickam and Y. Yamaguchi, “An Adaptive General Four-Component Scattering Power Decomposition With Unitary Transformation of Coherency Matrix (AG4U),” in IEEE Geoscience and Remote Sensing Letters, vol. 12, no. 10, pp. 2110-2114, Oct. 2015. DOI: 10.1109/LGRS.2015.2451369
Compact PolSAR Decompositions (S − Ω)
We have proposed a new decomposition technique for compact polarimetric (CP) synthetic aperture radar (SAR) data. In the proposed decomposition, the odd and the even bounce scattering powers obtained by combining the powers received in the opposite-sense circular (OC) polarization and the same-sense circular (SC) polarization transmitted with the polarized power fraction ( Ω ). The volume scattering power is obtained by combining the total power with the unpolarized power fraction. The parameter is a function of both the transmitting and the receiving ellipticities and orientations. These parameters thus provide a wider degree of freedom to accommodate a range of scattering mechanisms which are not reflected in the existing approaches.
Fig: Decomposition powers for the simulated (with AR = 0 dB and ψt=0) AIRSAR L-band Flevoland data set (a) Pauli RGB with a transect over a trihedral corner reflector. (b) SC (red) and OC (blue) powers with Ω≈0.7; (c) S − Ω and (d) m-χ decomposition.
The proposed method is applied on simulated CP-SAR data obtained from full-polarimetric E-SAR (Experimental Synthetic Aperture Radar) and AIRSAR (Airborne Synthetic Aperture Radar) L-band data sets. The proposed decomposition shows appreciable improvements in the scattering powers compared to the m-δ and the m-χ decompositions.A. Bhattacharya, S. De, A. Muhuri, M. Surendar, G. Venkataraman, A. K. Das, “A New Compact Polarimetric SAR Decomposition Technique”, Remote Sensing Letters, vol. 6, no. 12, pp. 914-923, Dec. 2015, DOI:10.1080/2150704X.2015.1088669.