Abstract
This article derives a new change detector for multivariate synthetic aperture radar (SAR) image time series (ITS). Classical statistical change detection methodologies based on covariance matrix analysis are usually built upon the Gaussian assumption, as well as an unstructured signal model. Both of these hypotheses may be inaccurate for high-dimension/resolution images, where the noise can be heterogeneous (non-Gaussian) and where the relevant signals usually lie in a low-dimensional subspace (low-rank structure). These two issues are tackled by proposing a new generalized likelihood ratio test based on a robust (compound Gaussian) low-rank (structured covariance matrix) model. The interest of the proposed detector is assessed on two SAR-ITS set from UAVSAR.
Highlights
T HE analysis of synthetic aperture radar (SAR) image time series (ITS) has become a popular topic of research since it has many practical applications for earth monitoring, such as disaster assessment, infrastructure monitoring, or land-cover analysis
An active topic of research addresses the development of reliable automatic change detection (CD) methodologies in order to efficiently process this large amount of data
Assuming Gaussian distributed samples, the CD can be performed by testing a change in the covariance matrix [6], [7]
Summary
T HE analysis of synthetic aperture radar (SAR) image time series (ITS) has become a popular topic of research since it has many practical applications for earth monitoring, such as disaster assessment, infrastructure monitoring, or land-cover analysis. Another example would be the case where only one polarization channel contains relevant information for change detection purpose In this scope, [28] proposed to extend the GLRT approach to test for the equality of the parameters of low-rank structured covariance matrices. To account for both issues, this article proposes a new CD method based on both robust and low-rank (LR) models: we derive a GLRT for CG distributed observations that have a low-rank structured covariance matrix. The formulation of this test involves nontrivial optimization problems for which we tailor a practical block-coordinate descent algorithm. Hp+,R denotes the set of p × p Hermitian positive semidefinite matrices of rank R. ∝ stands for “proportional to”. x ∼ CN (μ, Σ) is a complex-valued random Gaussian vector of mean μ and covariance matrix Σ
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have