Computer Engineering and Applications ›› 2022, Vol. 58 ›› Issue (1): 241-247.DOI: 10.3778/j.issn.1002-8331.2008-0285

• Graphics and Image Processing • Previous Articles     Next Articles

GPR High-Frequency Clutter Suppression Based on Nonnegative Matrix Factorization

MIAO Cui, YUAN Da, WANG Dongyu, LI Wensheng   

  1. Key Laboratory of Intelligent Information Processing in Universities of Shandong(Shandong Technology and Business University), Yantai, Shandong 264005, China
  • Online:2022-01-01 Published:2022-01-06

基于非负矩阵分解的GPR高频杂波抑制

苗翠,原达,王冬雨,李文生   

  1. 山东省高校智能信息处理重点实验室(山东工商学院),山东 烟台 264005

Abstract: Aiming at the problem of ground penetrating radar(ground penetrating radar, GPR) collecting data, high-frequency clutter will interfere with the automatic identification of underground targets. A nonnegative matrix decomposition method(probability nonnegative matrix factorization, PNMF) for GPR images based on variational bayes is proposed. This method uses a variational bayes model to decompose the base matrix and coefficient matrix of the nonnegative matrix. It performs approximate reasoning to obtain the low-rank matrix component representation of the clutter, and then separates the clutter from the image. In the experimental process, simulation and measured data are used for comparative analysis. The signal-to-noise ratio and visual quality results verify that PNMF has a good suppression effect on clutter and has good robustness.

Key words: ground penetrating radar(GPR), high-frequency clutter, variational Bayes, nonnegative matrix factorization

摘要: 针对探地雷达(ground penetrating radar,GPR)采集数据时会产生高频杂波影响地下目标自动识别的问题。提出了一种基于变分贝叶斯的GPR图像非负矩阵分解方法(probability nonnegative matrix factorization,PNMF)。该方法使用变分贝叶斯模型对非负矩阵分解的基矩阵和系数矩阵进行近似推理,得到杂波成分的低秩矩阵表示,进而将杂波从图像中分离出来。实验过程采用模拟和实测数据进行对比分析,通过信噪比和视觉质量结果验证了PNMF对杂波有较好的抑制作用,具有较好的鲁棒性。

关键词: 探地雷达, 高频杂波, 变分贝叶斯, 非负矩阵分解