使用Savitzky—Golay滤波器改进的位场ISVD算法

Improved Potential Field ISVD Algorithm Based on Savitzky—Golay Filter

结合空间域和频率域计算的ISVD(Integrated Second Vertical Derivative)位场垂向导数换算算法,相对单纯的空间域或频率域方法具有较高的稳定性,但是ISVD算法在高阶垂向导数换算过程中依然会在一定程度上受到噪声的干扰,随着换算阶数的增加,噪声干扰逐级增大。为了能够在压制噪声的同时很好地保持异常真实形态,将具有异常形态高保真特点Savitzky-Golay滤波器引入ISVD算法中,通过选择最优滤波参数,在各阶垂向导数的换算过程中能够对噪声进行逐级压制,同时维持了异常的真实信息。通过模型验证了Savitzky-Golay滤波对于ISVD算法的稳定性具有明显的改进效果。

ISVD(integrated second vertical derivative)is a vertical derivative calculating algorithm which combines the space domain and frequency domain calculation.Compared to the space domain or frequency domain method,ISVD owns higher stability,butit still be disturbed by the noise to some extent in the process of higher vertical derivative conversion,and the noise disturbance will sequentially augment as the conversion order grows.In order to suppress the noise and keep the anomaly pattern real,the Savitzky—Golay filter is introduced into the ISVD algorithm.By selecting the optimal filter parameters,the noise can be suppressedsequentially,meanwhile,the real morphology of the anomaly can be kept well.The model test demonstrated that the Savitzky—Golay filter can improve the stability of ISVD algorithm explicitly.