An overview of clutter mitigation methods for tomographic material inspection

Millimeter wave-based material inspection methods are capable of providing 3-D-tomographic images of composite materials with a sufficient resolution to detect, localize, and classify critical defects. The nondestructive and noncontact-based measurement methods' potential can only be utilized with a complete understanding of the signals propagation behavior. Such that the relevant signal processing is tailored to the respective application. A core challenge of these systems remains clutter mitigation. Due to the composite materials' structure, a significant component of the received signal is corrupted by internal reflections, manifesting as clutter. For this purpose, a variety of different clutter mitigation methods, which include: normalization, moving average and moving median filter, entropy-based filtering, time gating, spatial filtering, and singular value decomposition, are investigated. These methods are applied to tomographic data of glass fiber reinforced plastic components, acquired via a W-band imaging system.