Detection of Moving Ships Based on a Combination of Magnitude and Phase in Along-Track Interferometric SAR-Part II: Statistical Modeling and CFAR Detection

A new powerful constant false-alarm rate (CFAR) method is proposed for detecting moving ships in along-track interferometric synthetic aperture radar (ATI-SAR) images, based on the sea interferogram's magnitude and phase (SIMP) metric, presented in Part I of this paper. First, within the structure of the product model, a statistical model (hereafter simply termed Nq) for the square root of SIMP metrics of sea clutter was developed by assuming the radar cross-sectional (RCS) components of the return follow a commonly used gamma distribution. Moreover, the parameter estimators of the presented model were analytically derived by applying the well-known "method of log cumulants" (MoLC). Finally, the CFAR threshold of moving ship detection, using the proposed model was given analytically. The experimental results of measured ATI-SAR data by NASA/JPL's AirSAR show that the proposed method gives good performance.