Research on SAR images recognition based on ART2 neural network

Author

Ye, Xiaoming ; Gao, Wei ; Wang, Yi ; Hu, Xiaoguang

Author_Institution

Sch. of Autom. Sci. & Electr. Eng., Beihang Univ., Beijing, China

fYear

2012

fDate

18-20 July 2012

Firstpage

1888

Lastpage

1891

Abstract

ART2 is a kind of self-organizing neural network which is based on adaptive resonance theory. It carries out the recognition by using competive learning and self-steady mechanism, and can learn by itself in dynamic environment with noise and without supervision. Its learning process can recognize learned models fastly and be adapted to new unknown objects rapidly. SAR ATR (Synthetic Aperture Radar Automatic Target Recognition) approach based on PCA and ART2 neural network is proposed in this paper. It takes the principal components as sample features, and then ART2 neural network is used to recognize SAR images. Experimental results with MSTAR SAR data sets show a better performance of recognition and generalization.

Keywords

ART neural nets; image recognition; learning (artificial intelligence); principal component analysis; radar computing; radar imaging; radar target recognition; synthetic aperture radar; ART2 neural network; MSTAR SAR data sets; PCA; SAR ATR; SAR images recognition; adaptive resonance theory; competive learning; dynamic environment; learned models; learning process; principal components; self-organizing neural network; self-steady mechanism; synthetic aperture radar automatic target recognition approach; Feature extraction; Neural networks; Principal component analysis; Synthetic aperture radar; Target recognition; Training; Vectors; ART2 neural network; PCA; SAR; recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Industrial Electronics and Applications (ICIEA), 2012 7th IEEE Conference on

Conference_Location

Singapore

Print_ISBN

978-1-4577-2118-2

Type

conf

DOI

10.1109/ICIEA.2012.6361036

Filename

6361036