摘要
针对重力场反演中深度分辨率过低的问题,提出基于混合稀疏正则化联合反演。引入L1多尺度小波和全变差技术,使在剔除干扰异常及规避反演过度聚焦的同时,尽可能地保留深度方向上的分辨率;针对在混合正则化反演中预条件矩阵左乘(求逆)计算量过大的问题,构建新的深度权项实现基于最小分裂Bregman迭代算法快速三维密度反演成像。研究结果表明:本文反演方法可行、有效,具有较强的适应性。
Considering the depth resolution of gravity inversion is too low, a hyper-parameter regularization algorithm based on multi-scale L1 wavelet operator and total variation operator was proposed. Regarding the difficulty to calculate the inverse matrix of preconditioning matrix, a new preconditioner was introduced to carry out the hyper-parameter regularization inversion, which was solved by the minimum splitting Bregman iterative algorithm. The results show that the proposed method is vilid and effective, and has strong adaptability.
出处
《中南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2015年第5期1699-1706,共8页
Journal of Central South University:Science and Technology
基金
国家自然科学基金资助项目(41174061
41374120)~~
关键词
混合正则化
分裂Bregman迭代算法
深度加权矩阵
多尺度小波
全变差正则化
hyper-parameter regularization
splitting Bregman iterative algorithm
depth weighting matrix
multi-scale wavelet
total variation operator
