Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II

Research Article

A Novel Parameter Determination Method for Lq Regularization Based Sparse SAR Imaging

Download
135 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-73447-7_18,
        author={Jia-cheng Ni and Qun Zhang and Li Sun and Xian-jiao Liang},
        title={A Novel Parameter Determination Method for Lq Regularization Based Sparse SAR Imaging},
        proceedings={Machine Learning and Intelligent Communications. Second International Conference, MLICOM 2017, Weihai, China, August 5-6, 2017, Proceedings, Part II},
        proceedings_a={MLICOM},
        year={2018},
        month={2},
        keywords={SAR imaging Lq(0 < q < 1) regularization Regularization parameter determination},
        doi={10.1007/978-3-319-73447-7_18}
    }
    
  • Jia-cheng Ni
    Qun Zhang
    Li Sun
    Xian-jiao Liang
    Year: 2018
    A Novel Parameter Determination Method for Lq Regularization Based Sparse SAR Imaging
    MLICOM
    Springer
    DOI: 10.1007/978-3-319-73447-7_18
Jia-cheng Ni1,*, Qun Zhang, Li Sun1, Xian-jiao Liang2
  • 1: Air Force Engineering University
  • 2: Unit 95100 of PLA
*Contact email: littlenjc@sina.com

Abstract

Sparse SAR imaging based on Lq(0 < q < 1) regularization has become a hot issue in SAR imaging. However, it can be difficult to determine a suitable value of the regularization parameter. In this paper, we developed a novel adaptive regularization parameter determination method for Lq regularization based SAR imaging. On the basis that the noise type in SAR system is mostly additive Gaussian white noise, we present a method for determining the regularization parameter through evaluating the statistics of noise. The parameter is updated through validating the statistical properties of the reconstruction error residuals in a suitable Noise Confidence Region (NCR). The experiment results illustrate the validity of the proposed method.