Edges of potential field sources often denote geologic features and are therefore a problem of fundamental importance in geological interpretation. There are many different methods for locating the source edges, most of which are based on the derivatives of the potential field data. In general, these methods can be divided into two main groups, namely amplitude-based methods and phase-based methods. This study compares the performance of popular amplitude- and phase-based edge detection methods such as total horizontal gradient, analytic signal (also called total gradient), enhanced analytic signal, directional analytic signal, tilt derivative, theta map, horizontal tilt angle, hyperbolic tilt angle, and enhanced total horizontal gradient. The methods are tested on synthetic magnetic data, as well as synthetic gravity data with and without noise. Furthermore, the methods are also applied to real magnetic and gravity data sets from northeast Vietnam as practical examples. The results show that the enhanced total horizontal gradient performs better than other methods under almost all scenarios.

潜在场源的边缘通常表示地质特征，因此是地质解释中至关重要的问题。有许多不同的方法可以定位源极边缘，其中大多数方法都是基于势场数据的导数。通常，这些方法可以分为两个主要组，即基于幅度的方法和基于相位的方法。这项研究比较了流行的基于幅度和相位的边缘检测方法的性能，例如总水平梯度，分析信号（也称为总梯度），增强分析信号，方向分析信号，倾斜导数，θ贴图，水平倾斜角，双曲线倾斜角度和增强的总水平倾斜度。这些方法在合成磁数据以及有无噪声的合成重力数据上进行了测试。此外，这些方法还适用于越南东北部的实际磁和重力数据集，作为实例。结果表明，在几乎所有情况下，增强的总水平梯度的性能均优于其他方法。