Open AccessClassification of hybrid‐pol data using novel cross‐polarisation estimation approach
Author(s) -
Kumar A.,
Panigrahi R.K.
Publication year - 2018
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
ISSN - 1350-911X
DOI - 10.1049/el.2017.3855
Subject(s) - computer science , algorithm , classifier (uml) , polarimetry , cross validation , pixel , basis (linear algebra) , pattern recognition (psychology) , artificial intelligence , data mining , mathematics , optics , physics , scattering , geometry
A novel cross‐polarisation (S HV2 ) estimation approach is proposed, which can be used to calculate the real part of co‐polarisation cross‐product [ ℜ ( S HHS VV ⋆ ) ] value. The term ℜ ( S HHS VV ⋆ ) is well known and widely used in full‐polarimetric (full‐pol) decomposition techniques. Its role in the hybrid‐pol analysis is investigated. Using ℜ ( S HHS VV ⋆ ) , a maximum‐likelihood‐based classifier is proposed. For performance evaluation, the classification results obtained using the proposed algorithm is compared with that of conventional m − δ and modified m − χ techniques. Pixel‐wise comparisons of these three techniques are carried out over two different datasets, by considering full‐pol as the standard basis.