上海交通大学学报 ›› 2019, Vol. 53 ›› Issue (9): 1074-1083.doi: 10.16183/j.cnki.jsjtu.2019.09.009

• 学报(中文) • 上一篇    下一篇

基于形态反演的小尺度磁性目标三维重建方法

尹刚,张林   

  1. 中国空气动力研究与发展中心 高速空气动力研究所, 四川 绵阳 621000
  • 发布日期:2019-10-11
  • 作者简介:尹刚(1988-),男,安徽省亳州市人,博士,工程师,主要研究方向为弱信号测试与处理, E-mail: gang.gang88@163.com.
  • 基金资助:
    国防科研项目资助

Three-Dimensional Reconstruction Method for Small Scale Magnetic Target Based on Shape Inversion

YIN Gang,ZHANG Lin   

  1. High Speed Aerodynamics Research Institute, China Aerodynamics Research & Development Center, Mianyang 621000, Sichuan, China
  • Published:2019-10-11

摘要: 本文以小尺度磁性目标的三维重建方法为研究对象,首先利用磁梯度张量数据估计磁性目标的近似水平边界、垂向边界和磁化方向,在此基础上确定待反演空间的范围、网格化反演空间、选定磁性目标初始反演模型和待增长模块集;然后基于模块选择函数在待增长模块集中选择最优的增长模块,迭代更新磁性目标的反演模型和待增长模块集,直至达到迭代目标函数的终止阈值;最终得到的反演模型即为小尺度磁性目标的三维重建结果.模拟结果表明:小尺度磁性目标的三维重建结果与真实目标具有较高的相似性,可用于磁性目标的识别应用.

关键词: 磁性目标; 形态反演; 磁梯度张量; 三维重建; 倾斜角; 局部波数法

Abstract: Three-dimensional reconstruction method for small scale magnetic target based on shape inversion is studied in this paper. Firstly, the magnetic gradient tensor data is used to estimate the approximate horizontal boundary, vertical boundary and magnetization direction. On this basis, the range of inversion space is selected, the inversion space is gridded, the initial inversion model and the module sets to be increased are given. Then, based on the established module selection function, the optimal growth module is selected in the module sets to be increased, and the inversion model of magnetic target and the module sets to be increased are updated iteratively until the threshold of the iterative objective function is achieved. Finally, the inversion model is the three-dimensional reconstruction results of small scale magnetic target. Simulation result shows that the three-dimensional reconstruction results have high similarity with the real target, and it can be used in the identification applications of magnetic targets.

Key words: magnetic target; shape inversion; magnetic gradient tensor; three-dimensional reconstruction; tilt angle; local wavenumber method

中图分类号: