Blurness and focusing-defocusing for through the wall radar imaging with wall ambiguities

Abstract

One of the main problems facing through the wall radar imaging (TWRI) application is determining the wall characteristics at small or large stand-off distances. Wall ambiguities cause the image to change in intensity and shift in position. Several contrast based metrics have been introduced in the past to auto-focus the image in both radar and sonar applications. These metrics, although useful under high wall ambiguities, they are not sensitive to incremental errors. In this paper, we introduce new criteria that are sensitive to small errors in wall thickness and dielectric constant. These criteria use higher order statistics. The gamma correction, widely used in image processing, is used to preprocess the image to enhance the sensitivity to wall errors. Simulation results show that the new metrics assume large drop in values with small deviations from the correct wall parameters.