CFAR FOR HOMOGENEOUS PART OF HIGH-RESOLUTION IMAGERY

In high-resolution imaging, weak target pixel amplitudes may not be detected in the presence of clutter containing strong nonhomogeneities, when conventional approaches are used. We describe a constant false alarm rate (CFAR) approach that avoids the elimination of these significant target returns. The nonhomogeneous clutter as well as the weak target components are detected with this approach. The targets could then be discriminated from the nonhomogeneities by discrimination techniques. We show how the lower amplitude components of the background noise and homogeneous clutter (which have Rayleigh statistics) can be detected in the presence of strong nonhomogeneous clutter and targets. The average level of the homogeneous component is then determined using these lower amplitude components. This CFAR approach avoids having a CFAR on the strong nonhomogeneities as well as the homogeneous component. The avoidance is what gives us the ability to detect weak target pixel amplitudes.