Defining regions of interest for microwave imaging using near-field reflection data

Microwave imaging benefits from information on the internal structure of the object of interest. Previously, we developed a tool to define regions dominated by a particular material and tested this approach with 2-D simulations of breast models. In this paper, we apply the technique to experimental data with the goal to extract an object's internal structural information using radar-based techniques. Specifically, the technique extracts information from microwave reflection data in order to identify discontinuities in material properties. By combining observations from multiple antenna locations, this information is used to estimate regions of interest. The utility of the tool is demonstrated in practical scenarios where the data are generated experimentally from cylindrical models using an ultra-wideband sensor.