Target recognition study using SVM, ANNs and expert knowledge

Author

Shi, Guangzhi ; Hu, Junchuan ; Da, Lianglong ; Song, Rugang

Author_Institution

Dept. of Navig. & Commun., Navy Submarine Acad., Qingdao

fYear

2008

fDate

1-3 Sept. 2008

Firstpage

1507

Lastpage

1511

Abstract

An underwater acoustic target recognition system is researched. According to characteristic of the ship radiated-noise demodulation line spectrum feature and its training sample set, the target recognition system adopts four methods including expert system, neighbor method, SVM and RBF ANNs. And the target recognition system makes use of advantage of the four methods. Experiment results show that it has better recognition effect.

Keywords

demodulation; expert systems; radial basis function networks; ships; support vector machines; telecommunication computing; underwater acoustic communication; RBF ANN; SVM; expert system; neighbor method; ship radiated-noise demodulation; underwater acoustic target recognition system; Artificial intelligence; Artificial neural networks; Expert systems; Multi-layer neural network; Neural networks; Neurons; Support vector machine classification; Support vector machines; Target recognition; Underwater acoustics; Demodulation line spectrum feature; Expert system; RBF ANNs; SVM; Underwater acoustic target recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Automation and Logistics, 2008. ICAL 2008. IEEE International Conference on

Conference_Location

Qingdao

Print_ISBN

978-1-4244-2502-0

Electronic_ISBN

978-1-4244-2503-7

Type

conf

DOI

10.1109/ICAL.2008.4636392

Filename

4636392