Matrix Information Geometry Detector for Ship Target Detection in SAR Images

Ship target detection in Synthetic Aperture Radar (SAR) images faces significant challenges in complex backgrounds, particularly under low signal-to-clutter ratio conditions. While traditional constant false alarm rate (CFAR) method mainly depends on energy, their performance is often limited. To treat this issue, this paper proposes a Matrix Information Geometry (MIG) detector that leverages four geometric metrics and their mean on the symmetric positive definite (SPD) matrix manifold. By formulating the detection problem on the SPD matrix manifold, our approach establishes a mathematically rigorous framework that inherently captures the geometric relationships between targets and clutter. Experiments demonstrate the MIG detector can better suppress clutter, improve the discrimination between targets and clutter, and reduce the false alarm rate by more than 90%.