Target Detection and Texture Segmentation in Polarimetric SAR Images Using a Wavelet Frame: Theoretical Aspects

An extension to synthetic aperture radar (SAR) applications of a technique for target detection and texture segmentation using a wavelet frame is presented. Novel issues to be considered in the passage to radar imagery are the characterization of speckle statistics under a wavelet frame, and the influence of polarization states on texture measures. The work is presented in two parts. This paper deals with theoretical aspects, while application of the technique in a number of experiments will be addressed in a second installment. In particular, as to speckle, estimators that decouple the influence of speckle over texture are introduced and characterized by their expected value and variance. The response of the wavelet frame to discontinuities, an important issue in target detection problems, is addressed in terms of signal to speckle noise ratio. As to the extension to polarimetry, the notion of polarimetric texture is revisited, providing a theoretical model that explains dependencies of texture measures on the polarization states. To measure such effects in polarimetric SAR data a novel analytical tool is introduced called the Wavelet Polarimetric Signature (WASP). The tool encapsulates in graphical form the dependency of the wavelet variance on scale and polarization state. Finally an extension to polarimetric data of the discrete wavelet transform á trous algorithm is proposed. The theory exposed here underpins a method that has been proven successful and computationally attractive in a selected number of SAR thematic applications. It also sets the stage for the exploitation of novel target detection and textural segmentation capabilities based on polarimetric diversity.

DE GRANDI Gianfranco;  JONG-SEN Lee;  SCHULER Dale L.; 

2008-01-17

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC

JRC36486

0196-2892,   

http://ieeexplore.ieee.org/Xplore/defdeny.jsp?url=/iel5/36/4373350/04358832.pdf&tp=&arnumber=4358832&isnumber=4373350&code=2,    https://publications.jrc.ec.europa.eu/repository/handle/JRC36486,   

10.1109/TGRS.2007.905103,