Accession Number : ADA136455

Title :   An Adaptive, Noise Tolerant, Frequency Extrapolation Algorithm for Diffraction Corrected Ultrasound Tomography,

Corporate Author : UTAH UNIV SALT LAKE CITY

Personal Author(s) : Stenger,F ; Berggren,M J ; Johnson,S A ; Li,Y

PDF Url : ADA136455

Report Date : Dec 1983

Pagination or Media Count : 8

Abstract : In the authors previous work (1), a procedure was developed, based on algorithm for rational function frequency extrapolation and l sub 1 -norm averaging to circumvent both the effects of noise, and errors due to refracted curved paths, for solving the inverse scattering problem in ultrasonic imaging. In this paper, they reverse the order of the first two algorithms with respect to (1;) that is they first apply the l sub 1 averaging procedure and then carry out the rational function procedure, extrapolating to infinite frequency. An adaptive method for choosing the best rational function expansion is also employed. They describe the underlying ideas of the procedure, and illustrate examples of reconstructions in the presence and absence of Gaussian noise. These results are then compared with results of previous experiments. The resulting procedure works with a much smaller signal-to-noise ratio. An extension is suggested to image speed of sound and density. (Author)

Descriptors :   *Algorithms, *Tomography, *Inverse scattering, *Ultrasonics, *Image processing, Partial differential equations, Diffraction, Acoustic velocity, Gaussian noise, Signal to noise ratio, Problem solving

Subject Categories : Theoretical Mathematics
      Acoustics
      Optics

Distribution Statement : APPROVED FOR PUBLIC RELEASE