Title :
Radar object recognition by wavelet transform and neural network
Author :
Tkac, J. ; Spirko, S. ; Boka, Ladislav
Author_Institution :
Dept. of Radiolocation, Mil. Acad., Liptovsky Mikulas, Slovakia
Abstract :
This paper presents a method of radar object recognition by wavelet transform and neural network. Analysing range profiles by wavelet transform can show information about the tracked object more clearly. It also gives a method of constructing a feature vector for automatic recognition so that the dimension of the feature vector may be much smaller than the dimension of the primitive echo signal. A three-layer backpropagation neural network is used as recognisor. This problem is aimed at radar systems with the duration of the transmitted pulses longer than the size of the object
Keywords :
backpropagation; neural nets; object recognition; radar computing; radar imaging; radar target recognition; wavelet transforms; automatic object recognition; feature vector; neural network; primitive echo signal dimension; radar imaging system; radar object recognition; range profile analysis; three-layer backpropagation neural network; transmitted pulse duration; wavelet transform; Continuous wavelet transforms; Discrete wavelet transforms; Doppler radar; Feature extraction; Neural networks; Object recognition; Radar imaging; Radar signal processing; Wavelet coefficients; Wavelet transforms;
Conference_Titel :
Microwaves, Radar and Wireless Communications. 2000. MIKON-2000. 13th International Conference on
Conference_Location :
Wroclaw
Print_ISBN :
83-906662-3-5
DOI :
10.1109/MIKON.2000.913916
