Shape Identification of Target behind Wall Using Fourier Descriptor and Morphology

Abstract

Though-The-Wall Imaging Systems are promising candidates for search and rescue applications especially in the disaster areas, where victims are buried under collapsed walls. These applications require such systems to identify the shape of the target. Therefore, in this paper, a methodology to identify the shape of the target behind a wall has been presented. The developed methodology consists of data acquisition, 2D through-The-wall image formation of the target, feature extraction, and shape identification using neural network and image morphology technique. For this purpose, C-scan data in of 3.5GHz-5.5 GHz frequency range using SFCW radar was collected for different shapes of wooden and metallic targets. An effective training technique using image morphology has been used to increase the detection accuracy of the artificial neural network (ANN) model. From, the developed ANN model result the retrieved shape of the target was found to be in good agreement with original shape. © 2020 Indian Radio Science Society.