Novel Extension of ART2 in Surface Landmine Detection
The Adaptive Resonance Theory 2 (ART2) neural network architecture is extended to provide a fuzzy output value, which indicates the degree of familiarity of a new analogue input pattern to previously stored patterns in the long term memory of the network. The outputs of the multilayer perceptron and this modified ART2 provide an analogue value to a fuzzy rule-based fusion technique which also uses a processed polarisation resolved image as its third input. In real-time situations these two classifier outputs indicate the likelihood of a surface landmine target when presented with a number of multispectral and textural bands. Due to the modifications in ART2, this updated alternative architecture has improved real-time landmine detection capabilities.
KeywordsFuzzy Rule False Alarm Rate Input Pattern Neural Network Architecture Automatic Target Recognition
Unable to display preview. Download preview PDF.
- Jain, L.C. and Vemuri, R. (Editors), (1998), Industrial Applications of Neural Networks, CRC Press USA.Google Scholar
- Bower, M., Cloud, E., Duvoisin, H., Long, D., and Hackett, J. (1995), “Development of automatic target recognition for infrared sensor-based close range land mine detector,” Martin Marietta Technologies Inc. Tech. Report, P.O. BOX 555837, MP-040, Orlando FL 32812–5837.Google Scholar
- Carpenter, G.A. and Grossberg, S. (1987), Pattern Recognition by Self-organising Neural Networks, Academic Press Inc., pp. 399–410.Google Scholar
- Field, S., Davey, N., and Frank, R. (1996), “Using neural networks to analyse software complexity, ” Australian Journal of Intelligent Information Processing Systems, vol. 3, no. 3, pp. 14–31, Spring.Google Scholar
- Filippidis, A. and Jain, L.C. (1997), “Identity attribute information in a multi-band aerial photo image using neural networks to automatically detect targets, ” International Journal of Knowledge -based Intelligent Engineering Systems, vol. 1, no. 1., January.Google Scholar
- Carpenter, G. and Grossberg, S. (1987), Pattern Recognition by Self-Organising Neural Networks, Academic Press Inc., pp. 399–410.Google Scholar
- Fuelop, K. and Hall, J. (1996), Thermal Infrared landmine detection, Technical Report DSTO-TR-0295 AR-009–485, pp. 3–19, January.Google Scholar
- Stacy, N., Smith, R., and Nash, G. (1994), Automatic target recognition for the INGARRA airborne radar surveillance system, D.S.T.O. Microwave Radar Division Internal Report, pp. 1–12, August.Google Scholar
- Whitbread, P.J. (1992), Multispectral texture, Ph.D. Thesis, The University of Adelaide, October.Google Scholar
- Harlick, R.M., Shunmuhham, K., and Distein, I. (1973), “Textural features for image classification, ” IEEE Transactions, Man and Cybernetics, vol. SMC-3, no. 6, November.Google Scholar
- “Neuro-genetic optimiser version 32202, ” BioComp Systems Inc. 2871, 152nd. Avenue N.E. Redmond, WA 98052.Google Scholar
- Abdulghafour, M.B. (1992), Data fusion through fuzzy reasoning applied to feature extraction from multi-sensory images, Ph.D. Thesis, The University of Tennessee, Knoxville, pp. 41–96, December.Google Scholar