The registration of multi-band polarimetric synthetic aperture radar (SAR) images is a key problem for image fusion. This paper presents a feature based on the polarimetric scattering mechanism for registration using the scale invariant feature transform (SIFT) algorithm. The feature represents information for the main scattering component and the influence of the other scattering components distribution and has good consistency for multi-band polarimetric SAR images. Tests demonstrate that the feature difference between multi-band polarimetric SAR images is less than the total scattered power (Span) difference. The SIFT algorithm is then used to get more key points, more correct registration points, and better spatial distribution of the correct registration points. The registration efficiency is remarkably increased by this feature.
LI Qiaoliang, WANG Guoyou, LIU Jianguo, et al.Robust scale-invariant feature matching for remote sensing image registration[J]. IEEE Gieoscience and Remote Snsing Letter, 2009, 6(2): 287-291.
Lowe D G. Object recognition from local scale-invariant features [C]// Proceedings of 7th International Conference on Computer Vision. Kerkyra, Greece, 1999: 1150-1157.
Lowe D G. Distinctive image features from scale-invariant keypoints[J]. Internation Journal of Computer Vision, 2004, 60(2): 91-110.
Goncalves H, Corte-Real L, Gon-alves J A. Automatic image registration through image segmentation and SIFT[J]. IEEE Transaction on Gieoscience and Remote Sensing, 2011, 49(7): 2589-2600.
Schwind P, Suri S, Reinartz P, et al.Applicability of the SIFT operator to geometric SAR image registration[J]. International Journal of Remote Sensing, 2010, 31(8): 1959-1980.
Chureesampant K, Susaki J. Automatic GCP extraction of fully polarimetric SAR images[J]. IEEE Transaction on Gieoscience and Remote Sensing, 2014: 52(1): 137-148.
ZHOU Guangyi, CUI Yi, CHEN Yilun, et al.A new edge detection method of polarimetric SAR image using curvelet transform and the Duda operator[J]. Electronic Letter, 2010, 46(2): 167-169.
Borghys D, Perneeland C, Acheroy M. A hierarchical approach for registration of high-resolution polarimetric SAR images [C]// Proceedings of SPIE Image and Signal Processing for Remote Sensing VII. Toulouse, France, 2002: 11-22.
Suri S, Schwind P, Uhl J, et al.Modifications in the SIFT operator for effective SAR image matching[J].International Journal of Remote Sensing, 2010, 1(3): 243-256.
Cloude S R, Pottier E. A review of target decomposition theorems in radar polarimery[J]. IEEE Transaction on Gieoscience and Remote Sensing, 1996, 34(2): 498-518.
Witkin A P. Scale-space filtering [C]//Proceedings of International Joint Conference on Artificial Intelligence. Karlsruhe, Germany, 1983: 1019-1022.
Cloude S R, Pottier E. An entropy based classification scheme for land applications of polarimetric SAR[J]. IEEE Transaction on Gieoscience and Remote Sensing, 1997, 35(1): 68-78.