An application of the dimensionality reduction methods for contrast target detection in microwave imaging

Microwave imaging systems have the capability to see through an optically opaque surface. Ground penetrating radar is a very popular application of microwave imaging that detects buried objects, and it is useful for defense and civilian applications. The scattered signal from the hidden object depends on the dielectric contrast between the object and the surrounding environment. In this regard, if the target with higher dielectric constant is present in the vicinity of the lower one, the detection becomes a challenging task for the lower dielectric constant target. This paper proposes a technique using dimensionality reduction to detect the targets with dielectric contrast efficiently. The results highlight the detection of multiple targets with dielectric contrast in a single scanning area.