Floating small target detection in sea clutter via normalised Hurst exponent

An improved fractal-based detector using the normalised Hurst exponent is proposed to detect floating small targets in sea clutter. Analysis of the IPIX radar datasets shows that the Hurst exponent of sea clutter varies with sea state and the viewing geometry of the radar. Referring to the exploitation of the reference range cells in adaptive detection, the mean and variance of the Hurst exponent of sea clutter are first computed from clutter-only time series at the reference range cells around the cell under test (CUT). Then the Hurst exponent at the CUT is normalised by the estimated mean and variance to attain the normalised Hurst exponent. The experimental results of the IPIX radar datasets show that the fractal-based detector using the normalised Hurst exponent achieves better detection performance.