A belief theoretic approach for characterization of underwater munitions

Thanuka L. Wickramarathne, Shahriar Negahdaripour, Kamal Premaratne, Lisa N. Brisson, Pierre P. Beaujean

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Characterization, management and remediation of military munitions, especially in underwater environments, is a challenging task given all the technical and physical barriers. Optical cameras are better suited for identifying the physical shape of objects. But in underwater, low visibility almost prohibits the use of these cameras. Acoustic imaging is a good alternative to this, but the characteristics of imaging along with numerous artifacts of physical systems which are not easy to model, makes the object recognition task non-trivial. We explore here the possibility of exploiting the geometry of the object shadows for identification of objects itself. The inherited imperfections of the data and the numerous artifacts of sonar systems are counteracted via the use of a fusion algorithm which incorporates evidence from multiple perspectives. A Dempster-Shafer belief theoretic evidence updating scheme which is capable of modeling a wider variety of data imperfections is used for the fusion task. We illustrate the method via the use of real data obtained at a test site located in the Florida Atlantic University premises.

Original languageEnglish
Title of host publicationMTS/IEEE Seattle, OCEANS 2010
DOIs
StatePublished - Dec 1 2010
EventMTS/IEEE Seattle, OCEANS 2010 - Seattle, WA, United States
Duration: Sep 20 2010Sep 23 2010

Other

OtherMTS/IEEE Seattle, OCEANS 2010
CountryUnited States
CitySeattle, WA
Period9/20/109/23/10

Fingerprint

Fusion reactions
Cameras
Acoustic imaging
Defects
Sonar
Object recognition
Remediation
Visibility
Imaging techniques
Geometry

Keywords

  • Dempster-Shafer theory
  • Evidence fusion
  • Evidence updating
  • Sonar imaging
  • Underwater munitions

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Ocean Engineering

Cite this

Wickramarathne, T. L., Negahdaripour, S., Premaratne, K., Brisson, L. N., & Beaujean, P. P. (2010). A belief theoretic approach for characterization of underwater munitions. In MTS/IEEE Seattle, OCEANS 2010 [5664390] https://doi.org/10.1109/OCEANS.2010.5664390

A belief theoretic approach for characterization of underwater munitions. / Wickramarathne, Thanuka L.; Negahdaripour, Shahriar; Premaratne, Kamal; Brisson, Lisa N.; Beaujean, Pierre P.

MTS/IEEE Seattle, OCEANS 2010. 2010. 5664390.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wickramarathne, TL, Negahdaripour, S, Premaratne, K, Brisson, LN & Beaujean, PP 2010, A belief theoretic approach for characterization of underwater munitions. in MTS/IEEE Seattle, OCEANS 2010., 5664390, MTS/IEEE Seattle, OCEANS 2010, Seattle, WA, United States, 9/20/10. https://doi.org/10.1109/OCEANS.2010.5664390
Wickramarathne, Thanuka L. ; Negahdaripour, Shahriar ; Premaratne, Kamal ; Brisson, Lisa N. ; Beaujean, Pierre P. / A belief theoretic approach for characterization of underwater munitions. MTS/IEEE Seattle, OCEANS 2010. 2010.
@inproceedings{83c3e32bdfc74e4b854089c4dd6379a4,
title = "A belief theoretic approach for characterization of underwater munitions",
abstract = "Characterization, management and remediation of military munitions, especially in underwater environments, is a challenging task given all the technical and physical barriers. Optical cameras are better suited for identifying the physical shape of objects. But in underwater, low visibility almost prohibits the use of these cameras. Acoustic imaging is a good alternative to this, but the characteristics of imaging along with numerous artifacts of physical systems which are not easy to model, makes the object recognition task non-trivial. We explore here the possibility of exploiting the geometry of the object shadows for identification of objects itself. The inherited imperfections of the data and the numerous artifacts of sonar systems are counteracted via the use of a fusion algorithm which incorporates evidence from multiple perspectives. A Dempster-Shafer belief theoretic evidence updating scheme which is capable of modeling a wider variety of data imperfections is used for the fusion task. We illustrate the method via the use of real data obtained at a test site located in the Florida Atlantic University premises.",
keywords = "Dempster-Shafer theory, Evidence fusion, Evidence updating, Sonar imaging, Underwater munitions",
author = "Wickramarathne, {Thanuka L.} and Shahriar Negahdaripour and Kamal Premaratne and Brisson, {Lisa N.} and Beaujean, {Pierre P.}",
year = "2010",
month = "12",
day = "1",
doi = "10.1109/OCEANS.2010.5664390",
language = "English",
isbn = "9781424443321",
booktitle = "MTS/IEEE Seattle, OCEANS 2010",

}

TY - GEN

T1 - A belief theoretic approach for characterization of underwater munitions

AU - Wickramarathne, Thanuka L.

AU - Negahdaripour, Shahriar

AU - Premaratne, Kamal

AU - Brisson, Lisa N.

AU - Beaujean, Pierre P.

PY - 2010/12/1

Y1 - 2010/12/1

N2 - Characterization, management and remediation of military munitions, especially in underwater environments, is a challenging task given all the technical and physical barriers. Optical cameras are better suited for identifying the physical shape of objects. But in underwater, low visibility almost prohibits the use of these cameras. Acoustic imaging is a good alternative to this, but the characteristics of imaging along with numerous artifacts of physical systems which are not easy to model, makes the object recognition task non-trivial. We explore here the possibility of exploiting the geometry of the object shadows for identification of objects itself. The inherited imperfections of the data and the numerous artifacts of sonar systems are counteracted via the use of a fusion algorithm which incorporates evidence from multiple perspectives. A Dempster-Shafer belief theoretic evidence updating scheme which is capable of modeling a wider variety of data imperfections is used for the fusion task. We illustrate the method via the use of real data obtained at a test site located in the Florida Atlantic University premises.

AB - Characterization, management and remediation of military munitions, especially in underwater environments, is a challenging task given all the technical and physical barriers. Optical cameras are better suited for identifying the physical shape of objects. But in underwater, low visibility almost prohibits the use of these cameras. Acoustic imaging is a good alternative to this, but the characteristics of imaging along with numerous artifacts of physical systems which are not easy to model, makes the object recognition task non-trivial. We explore here the possibility of exploiting the geometry of the object shadows for identification of objects itself. The inherited imperfections of the data and the numerous artifacts of sonar systems are counteracted via the use of a fusion algorithm which incorporates evidence from multiple perspectives. A Dempster-Shafer belief theoretic evidence updating scheme which is capable of modeling a wider variety of data imperfections is used for the fusion task. We illustrate the method via the use of real data obtained at a test site located in the Florida Atlantic University premises.

KW - Dempster-Shafer theory

KW - Evidence fusion

KW - Evidence updating

KW - Sonar imaging

KW - Underwater munitions

UR - http://www.scopus.com/inward/record.url?scp=78651330443&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78651330443&partnerID=8YFLogxK

U2 - 10.1109/OCEANS.2010.5664390

DO - 10.1109/OCEANS.2010.5664390

M3 - Conference contribution

SN - 9781424443321

BT - MTS/IEEE Seattle, OCEANS 2010

ER -