Focusing late-time returns from elastic targets

Targets encountered during synthetic aperture sonar (SAS) surveys may exhibit elastic scattering behavior and re-radiate sound after initial interrogation. These re-radiated returns are often described as “late-time” energy, as they reach the sonar after the initial geometric returns. The range-gene...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Dalton, Kyle S., Blanford, Thomas E., Brown, Daniel C.
Format: Tagungsbericht
Sprache:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Targets encountered during synthetic aperture sonar (SAS) surveys may exhibit elastic scattering behavior and re-radiate sound after initial interrogation. These re-radiated returns are often described as “late-time” energy, as they reach the sonar after the initial geometric returns. The range-general image reconstruction methods commonly used in SAS surveys do not properly account for late-time returns, causing late-time components to appear out-of-focus and away from the target’s true location. This presentation will discuss the development of a late-time focused reconstruction algorithm and its application to sonar surveys for unexploded ordnance (UXO). The proposed late-time focused method allows for selective application of range-general and range-specific reconstruction and removes far-field assumptions, both of which are improvements over previously developed techniques. Starting with modeled data, then data from an in-air sonar, and finally in-water data from a fielded sonar system, this work will demonstrate the qualitative differences between range-general imagery and late-time focused imagery. When using late-time focused reconstruction, late-time returns from resonant targets appear better-defined, in a spatially-tighter region near the target. Future work will substantiate the observed qualitative improvements with numeric image quality metrics and focus on applications to automatic target recognition.
ISSN:1939-800X
DOI:10.1121/2.0001780