| 摘要: |
| 针对现有步进频率探地雷达一维成像方法分辨率低、速度慢的问题,提出了一种改进的基追踪去噪一维快速超分辨成像方法。该方法首先建立步进频率回波信号模型,通过FFT+FT细化探地雷达感兴趣的目标区域,以构建轻量的基矩阵；然后基于目标在频域的稀疏特性构建稀疏方程,并利用基追踪去噪方法求解稀疏方程,最后将解与雷克子波褶积得到一维超分辨像。仿真和实验结果表明,该方法在保证高分辨的同时精度高,具有较快的信号处理速度和良好的抗噪声性能,与基追踪去噪算法相比,运行速度提高了约10倍。 |
| 关键词: 步进频率信号 超分辨成像 探地雷达 改进的基追踪去噪 |
| DOI： |
| 分类号:TN957.51 |
| 基金项目:南宁市科学研究与技术开发计划项目（中央引导地方科技发展资金项目）(No.20231011)，产研计划项目(CYY-HT2023-JSJJ-0023)，国家自然科学基金（No.62361015）;广西八桂学者项目资助； |
|
| Stepped-Frequency Ground Penetrating Radar Fast Super-Resolution Imaging Method |
|
|
|
|
| Abstract: |
| In order to solve the problem that low resolution and slow speed in existing stepped-frequency ground penetrating radar one-dimensional imaging methods, an improved basis pursuit denoising one-dimensional fast super-resolution imaging method is proposed. This method first establishes a stepped-frequency echo signal model and refines the region of interest in the GPR using FFT+FT to construct a lightweight basis matrix. Then, based on the sparse characteristics of the target in the frequency domain, construct a sparse equation, utilize the basis pursuit denoising methods to solve the sparse equation, and finally convolve the solution with a Ricker wavelet to obtain a one-dimensional super-resolution image. Simulation and experimental results verify that this method achieves high resolution and high accuracy while ensuring fast signal processing speed and good noise resistance. Compared to the basis pursuit denoising algorithm, the running speed is improved by approximately 10 times. |
| Key words: stepped-frequency signal super-resolution image ground penetrating radar improved basis pursuit denoising |