利用水平与垂直导数的相关系数进行位场数据的边界识别

Edge Detection of Potential Filed Data Using Correlation Coefficients of Horizontal and Vertical Derivatives

为了解决现今边界识别方法中普遍存在的不能同时识别不同深度地质体边界的问题,采用相关系数法进行边界识别。该方法是利用总水平导数与垂直导数的相关系数来进行边界的识别。由于不同地质体边界处的水平与垂直导数的相关系数幅值几乎一样,因此该方法可以有效地同时识别出不同深度地质体的边界。理论模型试验表明:该方法能同时识别出不同深度地质体的边界,且与模型边界相吻合;将其应用于实际数据的处理中,该方法能识别出更多的地质细节。

In order to solve the existing fault of edge detection methods that cannot simultaneously identify the geological edges in different depth,this paper presented a new edge detection method,named as correlation coefficient,based on the correlation coefficients of general horizontal derivatives and vertical derivatives;the amplitudes at the edges of different geological bodies are almost the same,so this method can effectively identify the geological edges in different depth simultaneously.The test of theoretical model indicates that this method could detect the geological edges in different depth and the result is in accordance with model edge.We applied this method to real data,compared with other edge detection methods,this method could recognize more geologic details.