Bi-Phase Compound-Gaussian Mixture Model of Sea Clutter and Scene-Segmentation-Based Target Detection

In high-resolution maritime surveillance radars, sea clutter exhibits highly spatial heterogeneity due to modulation of long waves with wavelengths longer than the width one range cell. Compound-Gaussian model (CGM) fails characterize heterogeneous in both amplitude distribution and Doppler spectrum. this article, a bi-phase compound-Gaussian mixture (BP-CGMM) is proposed clutter. BP-CGMM, resolution cells are grouped into two disjoint sets, each set represented by CGM inverse Gamma-distributed texture. The spectral indicates that vectors at spatially adjacent share same speckle covariance matrix, while separated sets often have different matrices. BP-CGMM validated mass measured data. Moreover, under detection method based on batch test given detect sea-surface small targets, which composed scene segmentation, aid Bayesian threshold morphological filtering, adaptive generalized likelihood ratio linear-threshold detector (GLRT-LTD) separately set. verified data targets test. experimental results show it attains better performance GLRT-LTD