On the confidence level fusion of IR and forward-looking GPR

Tsaipei Wang; J. M. Keller; M. Busch; P. Gader; C. Hawkins; J. McElroy; K. C. Ho

doi:10.1117/12.666282

On the confidence level fusion of IR and forward-looking GPR

Tsaipei Wang^*, J. M. Keller, M. Busch, P. Gader, C. Hawkins, J. McElroy, K. C. Ho

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

We consider in this paper the improvement of side-attack mine detection by performing confidence level fusion with data collected from vehicle-mounted forward-looking IR and GPR (FL-GPR) sensors. The mine detection system is vehicle based, and has both IR and FL-GPR sensors mounted on the top of the vehicle. The IR images and FL-GPR data are captured as the vehicle moves forward. The detections from IR images are obtained from the Scale-Invariant Feature Transform (SIFT) and Morphological Shared-Weight Neural Networks (MSNN) depending on target characteristics, and those from FL-GPR are derived from the FL-GPR SAR images through object-tracking. Since the IR and FL-GPR alarms do not occur at the same location, the fusion process begins with each IR alarm and looks at the nearby FL-GPR alarms with confidences weighted by values that are inversely proportional to their distances to the IR alarm. The FL-GPR alarm with the highest weighted confidence is selected and combined with the IR confidence through geometric mean. An experimental dataset collected from a government test site is used for performance evaluation. At the highest Pd and comparing with IR only, fusing IR and FL-GPR yields a reduction of FAR by 26%. When the Hough transform is applied to reject the IR alarms that have irregular shapes, the fusion results provides a reduction of FAR by 35% at the highest Pd.

Original language	English
Title of host publication	Detection and Remediation Technologies for Mines and Minelike Targets XI
DOIs	https://doi.org/10.1117/12.666282
State	Published - 2006
Event	Detection and Remediation Technologies for Mines and Minelike Targets XI - Kissimmee, FL, United States Duration: 17 Apr 2006 → 21 Apr 2006

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	6217 II
ISSN (Print)	0277-786X

Conference

Conference	Detection and Remediation Technologies for Mines and Minelike Targets XI
Country/Territory	United States
City	Kissimmee, FL
Period	17/04/06 → 21/04/06

Keywords

Confidence level fusion
Forward-looking GPR
Forward-looking IR

Access to Document

10.1117/12.666282

Cite this

@inproceedings{2289a3743b614eda927f910a033497c5,

title = "On the confidence level fusion of IR and forward-looking GPR",

abstract = "We consider in this paper the improvement of side-attack mine detection by performing confidence level fusion with data collected from vehicle-mounted forward-looking IR and GPR (FL-GPR) sensors. The mine detection system is vehicle based, and has both IR and FL-GPR sensors mounted on the top of the vehicle. The IR images and FL-GPR data are captured as the vehicle moves forward. The detections from IR images are obtained from the Scale-Invariant Feature Transform (SIFT) and Morphological Shared-Weight Neural Networks (MSNN) depending on target characteristics, and those from FL-GPR are derived from the FL-GPR SAR images through object-tracking. Since the IR and FL-GPR alarms do not occur at the same location, the fusion process begins with each IR alarm and looks at the nearby FL-GPR alarms with confidences weighted by values that are inversely proportional to their distances to the IR alarm. The FL-GPR alarm with the highest weighted confidence is selected and combined with the IR confidence through geometric mean. An experimental dataset collected from a government test site is used for performance evaluation. At the highest Pd and comparing with IR only, fusing IR and FL-GPR yields a reduction of FAR by 26%. When the Hough transform is applied to reject the IR alarms that have irregular shapes, the fusion results provides a reduction of FAR by 35% at the highest Pd.",

keywords = "Confidence level fusion, Forward-looking GPR, Forward-looking IR",

author = "Tsaipei Wang and Keller, {J. M.} and M. Busch and P. Gader and C. Hawkins and J. McElroy and Ho, {K. C.}",

year = "2006",

doi = "10.1117/12.666282",

language = "English",

isbn = "081946273X",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

booktitle = "Detection and Remediation Technologies for Mines and Minelike Targets XI",

note = "Detection and Remediation Technologies for Mines and Minelike Targets XI ; Conference date: 17-04-2006 Through 21-04-2006",

}

Wang, T, Keller, JM, Busch, M, Gader, P, Hawkins, C, McElroy, J & Ho, KC 2006, On the confidence level fusion of IR and forward-looking GPR. in Detection and Remediation Technologies for Mines and Minelike Targets XI., 62172T, Proceedings of SPIE - The International Society for Optical Engineering, vol. 6217 II, Detection and Remediation Technologies for Mines and Minelike Targets XI, Kissimmee, FL, United States, 17/04/06. https://doi.org/10.1117/12.666282

On the confidence level fusion of IR and forward-looking GPR. / Wang, Tsaipei; Keller, J. M.; Busch, M. et al.
Detection and Remediation Technologies for Mines and Minelike Targets XI. 2006. 62172T (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 6217 II).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - On the confidence level fusion of IR and forward-looking GPR

AU - Wang, Tsaipei

AU - Keller, J. M.

AU - Busch, M.

AU - Gader, P.

AU - Hawkins, C.

AU - McElroy, J.

AU - Ho, K. C.

PY - 2006

Y1 - 2006

N2 - We consider in this paper the improvement of side-attack mine detection by performing confidence level fusion with data collected from vehicle-mounted forward-looking IR and GPR (FL-GPR) sensors. The mine detection system is vehicle based, and has both IR and FL-GPR sensors mounted on the top of the vehicle. The IR images and FL-GPR data are captured as the vehicle moves forward. The detections from IR images are obtained from the Scale-Invariant Feature Transform (SIFT) and Morphological Shared-Weight Neural Networks (MSNN) depending on target characteristics, and those from FL-GPR are derived from the FL-GPR SAR images through object-tracking. Since the IR and FL-GPR alarms do not occur at the same location, the fusion process begins with each IR alarm and looks at the nearby FL-GPR alarms with confidences weighted by values that are inversely proportional to their distances to the IR alarm. The FL-GPR alarm with the highest weighted confidence is selected and combined with the IR confidence through geometric mean. An experimental dataset collected from a government test site is used for performance evaluation. At the highest Pd and comparing with IR only, fusing IR and FL-GPR yields a reduction of FAR by 26%. When the Hough transform is applied to reject the IR alarms that have irregular shapes, the fusion results provides a reduction of FAR by 35% at the highest Pd.

AB - We consider in this paper the improvement of side-attack mine detection by performing confidence level fusion with data collected from vehicle-mounted forward-looking IR and GPR (FL-GPR) sensors. The mine detection system is vehicle based, and has both IR and FL-GPR sensors mounted on the top of the vehicle. The IR images and FL-GPR data are captured as the vehicle moves forward. The detections from IR images are obtained from the Scale-Invariant Feature Transform (SIFT) and Morphological Shared-Weight Neural Networks (MSNN) depending on target characteristics, and those from FL-GPR are derived from the FL-GPR SAR images through object-tracking. Since the IR and FL-GPR alarms do not occur at the same location, the fusion process begins with each IR alarm and looks at the nearby FL-GPR alarms with confidences weighted by values that are inversely proportional to their distances to the IR alarm. The FL-GPR alarm with the highest weighted confidence is selected and combined with the IR confidence through geometric mean. An experimental dataset collected from a government test site is used for performance evaluation. At the highest Pd and comparing with IR only, fusing IR and FL-GPR yields a reduction of FAR by 26%. When the Hough transform is applied to reject the IR alarms that have irregular shapes, the fusion results provides a reduction of FAR by 35% at the highest Pd.

KW - Confidence level fusion

KW - Forward-looking GPR

KW - Forward-looking IR

UR - http://www.scopus.com/inward/record.url?scp=33747374267&partnerID=8YFLogxK

U2 - 10.1117/12.666282

DO - 10.1117/12.666282

M3 - Conference contribution

AN - SCOPUS:33747374267

SN - 081946273X

SN - 9780819462732

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Detection and Remediation Technologies for Mines and Minelike Targets XI

T2 - Detection and Remediation Technologies for Mines and Minelike Targets XI

Y2 - 17 April 2006 through 21 April 2006

ER -

On the confidence level fusion of IR and forward-looking GPR

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this