This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Feature-Level and Decision-Level Fusion of Noncoincidently Sampled Sensors for Land Mine Detection
June 2001 (vol. 23 no. 6)
pp. 577-589

Abstract—We present and compare methods for feature-level (predetection) and decision-level (postdetection) fusion of multisensor data. This study emphasizes fusion techniques that are suitable for noncommensurate data sampled at noncoincident points. Decision-level fusion is most convenient for such data, but it is suboptimal in principle, since targets not detected by all sensors will not obtain the full benefits of fusion. A novel algorithm for feature-level fusion of noncommensurate, noncoincidently sampled data is described, in which a model is fitted to the sensor data and the model parameters are used as features. Formulations for both feature-level and decision-level fusion are described, along with some practical simplifications. A closed-form expression is available for feature-level fusion of normally distributed data and this expression is used with simulated data to study requirements for sample position accuracy in multisensor data. The performance of feature-level and decision-level fusion algorithms are compared for experimental data acquired by a metal detector, a ground-penetrating radar, and an infrared camera at a challenging test site containing surrogate mines. It is found that fusion of binary decisions does not perform significantly better than the best available sensor. The performance of feature-level fusion is significantly better than the individual sensors, as is decision-level fusion when detection confidence information is also available (“soft-decision” fusion).

[1] P.K. Varshney, Distributed Detection and Data Fusion. Springer-Verlag, 1997.
[2] B.V. Dasarathy, Decision Fusion. Los Alamitos, Calif.: IEEE CS, 1994.
[3] S. Chauduri, A. Crandall, and D. Reidy, “Multisensor Data Fusion for Mine Detection,” Proc. SPIE: Sensor Fusion III, R.C. Harney, ed., vol. 1306, pp. 187-203, 1990.
[4] B. Brusmark, A. Christiansen, P. Jägerbro, and A. Lauberts, “Combination of Ground Penetrating Radar and Metal Detector Data for Mine Detection,” Proc. Seventh Int'l Conf. Ground Penetrating Radar, 1998.
[5] G.A. Clark, S.K. Sengupta, R.J. Sherwood, J.E. Hernandez, M.R. Buhl, P.C. Schaich, R.J. Kane, M.J. Barth, and N.K. Del Grande, “Sensor Feature Fusion for Detecting Buried Objects,” Proc. SPIE: Underground and Obscured-Object Imaging and Detection, N.K. Del Grande, I. Cindrich, and P.B. Johnson, eds., vol. 1942, pp. 178-188, 1993.
[6] X. Miao, M.R. Azimi-Sadjadi, B. Tian, A.C. Dubey, and N.H. Witherspoon, “Detection of Mines and Minelike Targets Using Principal Component and Neural Methods,” IEEE Trans. Neural Networks, vol. 9, no. 3, pp. 454-463, May 1998.
[7] R.A. Weisenseel, W.C. Karl, D.A. Castanon, E. Miller, C. Rappaport, and C.A. Dimarzio, “Statistical Fusion of GPR and EMI Data,” Proc. SPIE: Detection and Remediation Technologies for Mines and Minelike Targets IV, A.C. Dubey, J.F. Harvey, J.T. Broach, and R.E. Dugan, eds., vol. 3710, pp. 1179-1187, 1999.
[8] E. den Breejen, K. Schutte, and F. Cremer, “Sensor Fusion for Anti-Personnel Landmine Detection, a Case Study,” Proc. SPIE: Detection and Remediation Technologies for Mines and Minelike Targets IV, A.C. Dubey, J.F. Harvey, J.T. Broach, and R.E. Dugan, Eds., vol. 3710, pp. 1235-1245, 1999.
[9] P.G. Johnson and P. Howard, “Performance Results of the EG&G Vehicle Mounted Mine Detector,” Proc. SPIE: Detection and Remediation Technologies for Mines and Minelike Targets IV, A.C. Dubey, J.F. Harvey, J.T. Broach, and R.E. Dugan, eds., vol. 3710, pp. 1149-1159, 1999.
[10] J.E. McFee, V.C. Aitken, R. Chesney, Y. Das, and K.L. Russell, “Multisensor Vehicle-Mounted Teleoperated Mine Detector with Data Fusion,” Proc. SPIE: Detection and Remediation Technologies for Mines and Mine-Like Targets, A.C. Dubey, J.F. Harvey, and J. Broach, eds., vol. 3392, pp. 1082-1093, 1998.
[11] A. Gunatilaka and B.A. Baertlein, “Comparison of Pre-Detection and Post-Detection Fusion for Mine Detection,” Proc. SPIE: Detection and Remediation Technologies for Mines and Mine-Like Targets, A.C. Dubey, J.F. Harvey, J.T. Broach, and R.E. Dugan, eds., vol. 3710, pp. 1212-1223, 1999,
[12] A. Gunatilaka and B.A. Baertlein, “A Subspace Decomposition Technique to Improve GPR Imaging of Anti-Personnel Mines,” Proc. SPIE: Detection and Remediation Technologies for Mines and Minelike Targets V, A.C. Dubey, J.F. Harvey, J.T. Broach, and R.E. Dugan, eds., vol. 4038, 2000.
[13] B.A. Baertlein, “Neural-Network Based Multi-Sensor Fusion for Mine Detection,” Technical Report N61331-93-C-0050, Ballena Systems Corporation, 1994.
[14] J.A. Marble, J.G. Ackenhusen, J.W. Wegrzyn, J. Mancuso, and C. Dwan, “Sensor Fusion Performance Gain for Buried Mine/uxo Detection Using GPR, EMI, and MAG Sensors,” Proc. SPIE: Detection and Remediation Technologies for Mines and Mine-Like Targets V, A.C. Dubey, J.F. Harvey, J.T. Broach, and R.E. Dugan, eds., vol. 4038, pp. 1473-1484, Apr. 2000.
[15] H.V. Poor, An Introduction to Signal Detection and Estimation. New York: Springer-Verlag, 1994.
[16] S.M. Kay, Fundamentals of Statistical Signal Processing: Estimation Theory. Englewood Cliffs, N.J.: Prentice-Hall, 1993.
[17] D.L. Hall, Mathematical Techniques in Multisensor Data Fusion. Boston: Artech House, 1992.
[18] Y. Das, J.E. McFee, J. Toews, and G.C. Stuart, “Analysis of an Electromagnetic Induction Detector for Real-Time Location of Buried Objects,” IEEE Trans. Geoscience Remote Sensing, vol. 28, no. 3, pp. 278-287, May 1990.
[19] S. Nag, L. PetersJr., I.J. Gupta, and C.C. Chen, “Ramp Response for the Detection of Anti-Personel Mines,” Proc. SPIE: Detection and Remediation Technologies for Mines and Minelike Targets IV, A.C. Dubey, J.F. Harvey, J.T. Broach, and R.E. Dugan, eds., vol. 3710, pp. 1313-1322, 1999.
[20] I.K. Sendur and B.A. Baertlein, “Techniques for Improving Buried Mine Detection in Thermal IR Imagery,” Proc. SPIE: Detection and Remediation Technologies for Mines and Minelike Targets IV, A.C. Dubey, J.F. Harvey, J.T. Broach, and R.E. Dugan, eds., vol. 3710, pp. 1272-1283, 1999.
[21] M.D. Richard and R.P. Lippmann, “Neural Network Classifiers Estimate Bayesian a Posteriori Probabilities,” Neural Computation, vol. 3, pp. 461-483, 1991.

Index Terms:
Land mines, sensor fusion, infrared, ground penetrating radar, metal detectors.
Citation:
Ajith H. Gunatilaka, Brian A. Baertlein, "Feature-Level and Decision-Level Fusion of Noncoincidently Sampled Sensors for Land Mine Detection," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no. 6, pp. 577-589, June 2001, doi:10.1109/34.927459
Usage of this product signifies your acceptance of the Terms of Use.