Moment-based method to statistically categorize rock outcrops based on their topographical features

Geomorphological formations on the ocean seafloor are spatially complex objects. Rock outcrops in particular can be spatially compact, and highly anisotropic in their large- and small-scale topographic structure, and represent a significant source of clutter for short-range active sonars operating in shallow water. This paper describes a physical moment-based statistical method for categorizing rock outcrops based on their topographical characteristics, in particular isolating its facet-like features. The correspondence of these features to clutter-generating objects is then substantiated for a short-range, high-frequency shallow-water scenario, where the ensonified area is a fraction of outcrop size, via comparisons with high-fidelity acoustic backscattering predictions.