Reconstruction of Ultrasound Reflector Images from Incomplete Data Using the Compressive Sensing Method

Abstract

The article investigates the possibility of reconstructing a reflector image using the compressive sensing (CS) method based on an incomplete set of echo signals measured by an antenna array in double or triple scanning mode. For comparison, we also considered reflector reconstruction methods used in ultrasonic testing (UT): the correlation, combined SAFT (C-SAFT), and maximum entropy (ME) methods. The last method allows reconstruction of superresolution images from an incomplete set of measured echo signals. Numerical and model experiments have demonstrated the possibility of reconstructing a reflector image with superresolution for a significant decrease in the amount of data used. Reconstructed CS images were compared with images reconstructed by other methods.