RESEARCH ON SAR IMAGE TARGET RECOGNITION ALGORITHM BASED ON NSCT+PNN

被引：0

作者：

Li, Dong ^{[1
]}

Gao, Binwen ^{[2
]}

Li, Ye ^{[3
]}

机构：

[1] Inner Mongolia Univ Technol, Coll Informat Engn, Hohhot, Peoples R China

[2] Inner Mongolia Power, Hohhot, Peoples R China

[3] Hohhot Vocat Coll, Hohhot, Peoples R China

来源：

UNIVERSITY POLITEHNICA OF BUCHAREST SCIENTIFIC BULLETIN SERIES C-ELECTRICAL ENGINEERING AND COMPUTER SCIENCE | 2020年 / 82卷 / 04期

关键词：

SAR; NSCT; PNN; Feature extraction; Target recognition;

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

Based on Synthesis Aperture Radar (SAR) image preprocessing, this paper proposes a SAR image target recognition method based on non-subsampled Contourlet Transform (NSCT) for feature extraction and Probabilistic Neural Network (PNN) for target recognition. On the basis of studying the denoising of multi-scale geometric analysis theory, the low frequency components and high-frequency components of SAR image samples are extracted by NSCT respectively. The target eigenvectors are obtained by PNN. This method was used to extract and identify the three types of military targets in the MSTAR database. The recognition rate was 90.7%. The results show that the SNCT+PNN method can effectively achieve SAR target recognition and obtain a higher target correct recognition rate.

引用

页码：173 / 184

页数：12