Synthetic Aperture Sonar Image Statistics

Open Access
Johnson, Shawn Fredrick
Graduate Program:
Doctor of Philosophy
Document Type:
Date of Defense:
January 30, 2009
Committee Members:
  • Dr Anthony Lyons, Dissertation Advisor
  • Anthony P Lyons, Committee Chair
  • Karl Martin Reichard, Committee Member
  • Ram Mohan Narayanan, Committee Member
  • Timothy Joseph Kane, Committee Member
  • resolution
  • image statistics
  • SAS
  • Synthetic Aperture Sonar
  • K distribution
  • multipath
Synthetic Aperture Sonar (SAS) systems are capable of producing photograph quality seafloor imagery using a lower frequency than other systems of comparable resolution. However, as with other high-resolution sonar systems, SAS imagery is often characterized by heavy-tailed amplitude distributions which may adversely affect target detection systems. The constant cross-range resolution with respect to range that results from the synthetic aperture formation process provides a unique opportunity to improve our understanding of system and environment interactions, which is essential for accurate performance prediction. This research focused on the impact of multipath contamination and the impact of resolution on image statistics, accomplished through analyses of data collected during at-sea experiments, analytical modeling, and development of numerical simulations. Multipath contamination was shown to have an appreciable impact on image statistics at ranges greater than the water depth and when the levels of the contributing multipath are within 10 dB of the direct path, reducing the image amplitude distribution tails while also degrading image clarity. Image statistics were shown to depend strongly upon both system resolution and orientation to seafloor features such as sand ripples. This work contributes to improving detection systems by aiding understanding of the influences of background (i.e. non-target) image statistics.