Immune response inspired spatial-temporal target detection algorithms with CNN-UM

• Published in: International journal of circuit theory and applications Vol. 34; no. 1; pp. 21 - 47
• Main Authors: Cserey, György, Falus, András, Roska, Tamás
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01.01.2006
• Subjects: analogic algorithms; artificial immune systems; cellular non-linear/neural networks (CNN); CNN Universal Machine
• ISSN: 0098-9886; 1097-007X
• DOI: 10.1002/cta.341

In this paper we show that, similar to the nervous system and the genetic system, the immune system provides a prototype for a ‘computing mechanism.’ We are presenting an immune response inspired algorithmic framework for spatial–temporal target detection applications using CNN technology (IEEE Trans. Circuits Syst. II 1993; 40(3):163–173; Foundations and Applications. Cambridge University Press: Cambridge, 2002). Unlike most analogic CNN algorithms (IEEE Trans. Circuits Syst. 1988; 35(10):1257–1290; Foundations and Applications. Cambridge University Press: Cambridge, 2002) here we will detect various targets by using a plethora of templates. These algorithms can be implemented successfully only by using a computer upon which thousands of elementary, fully parallel spatial–temporal actions can be implemented in real time. In our tests the results show a statistically complete success rate, and we are presenting a special example of recognizing dynamic objects. Results from tests in a 3D virtual world with different terrain textures are also reported to demonstrate that the system can detect unknown patterns and dynamical changes in image sequences. Applications of the system include in explorer systems for terrain surveillance. Copyright © 2006 John Wiley & Sons, Ltd.