The Hotelling-Lawley trace statistic for change detection in polarimetric SAR data under the complex Wishart distribution
In this paper we propose a new test statistic for unsupervised changedetectioninpolarimetricsyntheticapertureradar(PolSAR) data. We work with multilook complex (MLC) covariance matrix data, whose underlying model is assumed to be the scaled complex Wishart distribution. We use the complex kind Hotelling-Lawley (HL) trace statistic for measuring the similarity of two covariance matrices. The sampling distribution of the HL trace is approximated by a Fisher-Snedecor distribution,which is used to deﬁne the signiﬁcance level of a constant false alarm rate change detector. The performance of the proposed method is tested on simulated and real PolSAR data sets and compared to the likelihood ratio test statistic.
Embargo of 24 months from date of publishing on accepted manuscript version. Link to publisher's version:https://doi.org/10.1109/IGARSS.2013.6723750