Accession Number:

ADA224358

Title:

Modeling and Reconstruction Algorithms for Detection, Location, and Identification of Subsurface Anomalies

Descriptive Note:

Final rept. 15 Apr 1985-14 Apr 1990

Corporate Author:

ARIZONA STATE UNIV TEMPE

Personal Author(s):

Report Date:

1990-06-01

Pagination or Media Count:

293.0

Abstract:

Techniques for modeling and reconstructing images of subsurface anomalies, in a hole-to-hole system configuration, have been investigated. The region under examination is scanned by transmitting electromagnetic rays through the region. Numerical algorithms have been developed for solving the reconstruction problem. The singular value decomposition method has played a key role in fulfilling this goal. In particular, the conjugate gradient algorithms is used with the gradient projection method to obtain a highly stable and efficient technique which is able to incorporate inequality constraints. These inequality constraints represent one type of priori information that can be used ot yield better solution to detect, locate, and identify subsurface anomalies. The method of weighted least squares is used to include a priori information into the solution. Two ways of explicity adding this information into the reconstruction process are given, and their effectiveness to reducing additive noise is demonstrated. To obtain images of subsurface anomalies which are the most representative of the actual cross sections being investigated, both continuous wave and time-of-flight measurements are utilized.

Subject Categories:

  • Theoretical Mathematics
  • Target Direction, Range and Position Finding

Distribution Statement:

APPROVED FOR PUBLIC RELEASE