Title :
Hybrid neural networks for automatic target recognition
Author :
Waldemark, Joakim ; Becanovic, V. ; Lindblad, Th ; Lindsey, C.S.
Author_Institution :
Dept. of Phys., R. Inst. of Technol., Stockholm, Sweden
Abstract :
The paper presents a hybrid neural network system for automatic target recognition, or ATR. The ATR system uses a hybrid of a biological inspired neural net called the Pulse Coupled Neural Net, PCNN, and traditional feedforward neural nets. The PCNN is an iterative neural network in which, for example, a grey scale input image results in a 1D time signal invariant to rotation, scale and translation alternations. The PCNN can also extract edges, perform object segmentation and extract texture information. The PCNN pre-processor generates a 1D time signal that is input to a feedforward pattern recognition net
Keywords :
edge detection; feedforward neural nets; image segmentation; image texture; multilayer perceptrons; object recognition; 1D time signal; Pulse Coupled Neural Net; automatic target recognition; biological inspired neural net; edge extraction; feedforward neural nets; feedforward pattern recognition network; grey scale input image; hybrid neural network system; iterative neural network; object segmentation; pre-processor; rotation invariance; scale invariance; texture information; translation invariance; Data mining; Feedforward neural networks; Feeds; Image segmentation; Neural networks; Object segmentation; Pattern recognition; Physics; Signal generators; Target recognition;
Conference_Titel :
Systems, Man, and Cybernetics, 1997. Computational Cybernetics and Simulation., 1997 IEEE International Conference on
Print_ISBN :
0-7803-4053-1
DOI :
10.1109/ICSMC.1997.633300
