摘要
针对不均匀散射体重构问题,提出了一种基于Born近似下的贝叶斯压缩感知微波成像方法。在一阶Born近似框架下,基于电场积分方程并对成像区域网格离散建立稀疏感知模型,然后构造基于高斯先验的贝叶斯概率密度函数,并利用相关向量机方法对目标函数进行优化求解,最终实现对目标的微波成像。通过对多像素单目标、不均匀单目标、不均匀多目标等的微波成像研究并考虑了噪声影响,数值算例结果表明基于高斯先验的贝叶斯压缩感知方法重构结果要优于共轭梯度迭代算法和正交匹配追踪压缩感知重构算法,验证了文中方法的有效性和鲁棒性。
For the reconstruction problem of inhomogeneous scatterers,a Bayesian compressive sensing microwave imaging method based on Born approximation is proposed.In the framework of first-order Born approximation,a sparse sensing model is established based on the electric field integral equation and the mesh discretization in the imaging region.Then a Bayesian probability density function based on Gaussian Prior is constructed,and the objective function is optimized by using the relevant vector machine method.Finally,the microwave imaging of the target is realized.By studying the microwave imaging of multi-pixel single target,non-uniform single target and non-uniform multi-target and by considering the noise impact in the numerical example,the result is obtained and shows that the reconstruction result of Bayesian compressive sensing method based on Gaussian Prior is better than that of conjugate gradient iterative algorithm and orthogonal matching pursuit compressive sensing reconstruction algorithm,thus it verifies the effectiveness and robustness of the proposed method.
作者
于士奇
张清河
覃琴
王习东
YU Shi-qi;ZHANG Qing-he;QIN Qin;WANG Xi-dong(School of Computer and Information,Three Gorges University,Yichang 443002,China)
出处
《微波学报》
CSCD
北大核心
2020年第3期49-54,共6页
Journal of Microwaves
基金
国家自然科学基金(61179025,61771008)
湖北省自然科学基金(Z2018334/2018CKB914)。
关键词
微波成像
不均匀目标
Born近似
贝叶斯压缩感知
共轭梯度
microwave imaging
non-uniform target
Born approximation
Bayesian compressive sensing
conjugate gradient
