探地雷达子波确定性稀疏脉冲反褶积技术  

Deterministic-wavelet sparse spike deconvolution technique for ground penetrating radar data

作　　者：刘钰 石战结 王帮兵[1] 王征宇[3] 余天祥[1] LIU Yu;SHI Zhan-jie;WANG Bang-bing;Wang Zheng-Yu;YU Tian-xiang(School of Earth Sciences,Zhejiang University,Hangzhou 310027,China;Institute of Cultural Heritage,Zhejiang University,Hangzhou 310058,China;Hangzhou Institute of cultural relics and Archaeology,Hangzhou 310008,China)

机构地区：[1]浙江大学地球科学学院,浙江杭州310027 [2]浙江大学文化遗产研究院,浙江杭州310058 [3]杭州市文物考古研究所,浙江杭州310008

出　　处：《浙江大学学报（工学版）》2018年第9期1828-1836,共9页Journal of Zhejiang University：Engineering Science

基　　金：国家社会科学基金重大招标资助项目(13&ZD192);国家自然科学基金资助项目(41676181);中央高校基本科研业务费专项资助项目(ZJUSK2016007;2018QNA3013);浙江省文物保护科技资助项目(2013011;2015019;2017013);中石化石油工程地球物理有限公司胜利分公司科技资助项目(10200274-17-ZC0613-0018)

摘　　要：为了提升探地雷达数据的纵向分辨率,并解决传统反褶积方法对噪声敏感的问题,提出一种针对探地雷达信号的稀疏脉冲反褶积处理方法.通过试验获取子波信息,基于确定性的子波,采取正则化的反演方法,从信号中获得主要的反射系数.通过2个数值试验以及1个实际案例,对比分析处理前后的单道信号、剖面、频谱等情况,并与传统反褶积方法的结果做比较,结果表明:该方法相对于传统方法能较大地改善分辨率,并且受噪声影响较小,得到的结果成像效果更好.该方法经验证是一项有效、可行的探地雷达信号处理技术.A sparse spike deconvolution method for GPR signals was put forward in order to improve the vertical resolution of ground penetrating radar(GPR)data and solve the noise-sensitive problem of traditional deconvolution technique.Firstly the wavelet was acquired by experiment.Then a constraint inversion method using the deterministic wavelet was carried out to obtain the dominating reflectivity series from signals.The signals of single channel,profiles,spectrums,etc.before and after the process were analyzed through two numerical simulation experiments,as well as a practical case.Results were compared with that of the traditional deconvolution.The analyzing and comparing results show that the method can improve the resolution very well and is less affected by noise,which can provide a better image of the target.Our method is proved to be an effective and feasible GPR signal processing technique.

关 键 词：探地雷达(GPR) 子波 反褶积 反演 高分辨率 

分 类 号：P631[天文地球—地质矿产勘探]