This chapter has dealt with the problem of covariance matrix classification in PolSAR images on the base of the special structures assumed under symmetrical properties of the returns associated to the pixels under test. In particular, the chapter has focused on both homogeneous and heterogeneous SAR images' classification, including a description of the symmetry classification within the PolInSAR imagery. For all the described frameworks, the problem has been formulated as a multiple hypothesis test comprising both nested and non-nested hypotheses. For this reason, it has been solved by resorting to the well-known MOS rules to overcome the limitations of the classic GML approach. Results conducted on both simulated and L-band real-recorded PolSAR data have proven the effectiveness of the described methodologies, thus paving the way for further applications, e.g., as a preliminary step of a more sophisticated two-stage algorithm aimed at, for instance, classifying the scene.

Classification of covariance symmetries in full-polarimetric SAR images

Pallotta L.
;
2022-01-01

Abstract

This chapter has dealt with the problem of covariance matrix classification in PolSAR images on the base of the special structures assumed under symmetrical properties of the returns associated to the pixels under test. In particular, the chapter has focused on both homogeneous and heterogeneous SAR images' classification, including a description of the symmetry classification within the PolInSAR imagery. For all the described frameworks, the problem has been formulated as a multiple hypothesis test comprising both nested and non-nested hypotheses. For this reason, it has been solved by resorting to the well-known MOS rules to overcome the limitations of the classic GML approach. Results conducted on both simulated and L-band real-recorded PolSAR data have proven the effectiveness of the described methodologies, thus paving the way for further applications, e.g., as a preliminary step of a more sophisticated two-stage algorithm aimed at, for instance, classifying the scene.
2022
978-1-83953-402-7
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11563/163974
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? ND
social impact