J4 ›› 2014, Vol. 41 ›› Issue (4): 13-19.doi: 10.3969/j.issn.1001-2400.2014.04.003

• Original Articles • Previous Articles     Next Articles

Novel feature extraction method for synthetic aperture radar targets

DUAN Jia;ZHANG Lei;XING Mengdao;LIANG Yi   

  1. (National Key Lab. of Radar Signal Processing, Xidian Univ., Xi'an  710071, China)
  • Received:2013-06-18 Online:2014-08-20 Published:2014-09-25
  • Contact: DUAN Jia E-mail:bifiduan119@126.com

Abstract:

In order to extract geometric and electromagnetic features of targets from the radar echo, a novel parameter estimation algorithm for attributed scattering center decomposition and extraction is presented, which is steady and fast. By simplifying the model with consideration of real synthetic aperture radar working conditions, a simplified model is deduced. It is used to construct the base to represent the radar echo. The RELAX algorithm is utilized to determine the model parameter in an iterative way, with fast Fourier transform used for acceleration. Moreover, the maximum searching intervals and ranges are deduced in literature. Both simulated data and real measured data illustrate that our method is not only precise but also robust. Moreover, through the comparison with traditional methods, the efficiency has been improved effectively.

Key words: feature extraction, attributed scattering centers, parameter estimation, synthetic aperture radar