信号处理2025，Vol.41Issue(12)：1874-1891,18.DOI:10.12466/xhcl.2025.12.002
狄利克雷过程驱动的高斯混合先验贝叶斯学习SAR成像
Dirichlet Random Process-driven Bayesian Learning for Synthetic Aperture Radar Imagery via Gaussian Mixture Prior
摘要
Abstract
In high-resolution synthetic aperture radar(SAR)imaging,the existing prior distributions derived via statisti-cal methods are typically single and static.Consequently,the outcome of Bayesian inference depends on the specifics of any given problem.Therefore,existing models cannot solve problems with complex priors.Therefore,conventional methods fail to model detailed and refined features,which leads to incomplete retention of structural features of imaging targets and loss of weak scattering points.To address these issues,an SAR imaging algorithm based on Bayesian learn-ing is proposed,utilizing a Dirichlet process-driven Gaussian mixture prior(DPGMP-Bayes).Compared with conven-tional Bayesian modeling with random variables,statistical modeling methods that incorporate stochastic processes can model uncertainty with more flexibility.The Dirichlet process(DP)was employed to adaptively model the mixing weights of a Gaussian mixture model(GMM).This approach further optimized the modeling process of the GMM in dy-namically fitting complex prior distributions and achieved refined modeling of target features.Within the hierarchical Bayesian framework,the variational Bayes expectation maximization(VB-EM)algorithm was applied to adaptively in-fer hyperparameters.This technique enabled autonomous approximate inference of the posterior distribution,resulting in high-resolution imaging.Simulated and measured SAR data were used to compare the proposed approach with conven-tional imaging algorithms,and the results of qualitative and quantitative analyses validated that the proposed algorithm functioned as intended and exhibited superior performance compared with existing methods.关键词
合成孔径雷达成像/贝叶斯统计学习/高斯混合模型/狄利克雷过程/变分贝叶斯期望最大化Key words
synthetic aperture radar imaging/Bayesian learning/Gaussian mixture model/Dirichlet process/varia-tional Bayesian expectation maximization分类
信息技术与安全科学引用本文复制引用
杨磊,张泽楠,孙铭,樊后荣,孙鹏,周松..狄利克雷过程驱动的高斯混合先验贝叶斯学习SAR成像[J].信号处理,2025,41(12):1874-1891,18.基金项目
国家自然科学基金项目(62271487) （62271487）
江西省自然科学基金项目(20224ACB202003,20232ACB212003) The Natural Science Foundation of China(62271487) （20224ACB202003,20232ACB212003）
Natural Science Foundation of Jiangxi Province(20224ACB202003,20232ACB212003) （20224ACB202003,20232ACB212003）
