基于多结构元素形态滤波的大地电磁去噪

THE NOISE SUPPRESSION OF MAGNETOTELLURIC SOUNDING DATA BASED ON MULTIPLE STRUCTURAL ELEMENTS MATHEMATICAL MORPHOLOGY FILTERING

数学形态滤波是根据形态学原理创建的一种新型滤波方法,目前在信号处理的诸多领域得到成功运用。本文从形态滤波的基本原理出发,引入了一种多结构元素组成的滤波器用于大地电磁噪声消除。通过数值模拟展示了多结构元素滤波器的滤波效果,表明多结构元素滤波器在小尺度的范围内也有较好的滤波性能。同时,实测大地电磁信号的重构曲线和其视电阻率曲线显示多结构元素形态滤波对大地电磁噪声中的大尺度干扰噪声,如方波噪声和三角波噪声等,有较好的抑制作用,说明多结构元素形态滤波在大地电磁去噪中有较好的应用价值。

Mathematical morphological filtering, which is a new type of filtering method based on morphological principle, has been ap-plied successfully in various signal processing problems. Based on the principles of the morphological filtering, the authors introduced a new filter structure for noise cancellation characterized by the multiple structural elements to improve SNR （ Signal to Noise Ratio） of MT data in this paper. With numerical simulation, the filtering effect of the multiple structural elements mathematical morphology filte-ring is demonstrated,which indicates that the multiple structural elements mathematical morphology filtering has a good performance in supressing the noise at a small scale. The results of curve reconstruction of the real magnetotelluric sounding data and the apparent re-sistance curve show that the generalized morphological filter can suppress the large-scale noise such as square wave and triangular wave. It is shown that the multiple structural elements morphological filtering has a good application value in the noise suppression of magnetotelluric sounding data.