位场边缘识别方法研究进展

Some advances in the edge recognition of the potential field

研究地质体的边缘位置是重、磁位场数据解释永恒的主题,也是其优势.最近几年,国内外利用重、磁位场进行地质体边缘识别研究的文章明显增多,但没有作者系统整理和对比各方法的优点和缺点,给使用者带来诸多不便.本文首先将现有重、磁位场边缘识别方法分为数理统计、数值计算和其他三大类,并概述了各类方法的研究现状;之后较详细总结了数值计算类中垂向导数、总水平导数、解析信号振幅、倾斜角、θ图这5种基本的边缘识别方法以及在这些基本方法之上发展起来的诸如倾斜角总水平导数、增强解析信号振幅等方法的研究历史和应用效果;并用理论模型对比了几种主要边缘识别方法的识别效果.通过以上总结、对比和分析,指出了重、磁位场边缘识别方法使用中需要注意的问题以及将来的研究重点及发展方向.

The study on the edge location of the geological body is the eternal topic of the potential data interpretation, and this is also its advantage. In recent years, the articles about edge recognition using potential data are increasing apparently both at home and abroad. However, no author systematically arranged and compared advantages and disadvantages of each method. This brings much inconvenience to users. First of all, the methods of edge recognition are classified into three kinds of methods. They are Mathematical Statistics, Numerical Calculation and Other Methods, meanwhile the current situation of these methods have been outlined. Additionally, we summarized the fundamental methods of edge detection which are the vertical derivative （VDR）, total horizontal derivative（THDR）, analytical signal amplitude（ASM）, tilt angle（TA） and θ map（Theta Map）. We also described the researching history and the applied effect of TA_THDR（Total Horizontal Derivative of the Tilt Angle） which have been developed based on the fundamental methods of edge recognition. Based on this, we have contrasted the effects of the main methods of edge recognition through the synthetic model. Through summary, contrast and analysis, we have presented some problems that should be paid attention in edge recognition and have suggested the future research emphasis and developing direction.