Spike noise reduction method of TEM data based on stationary wavelet transform

Transient electromagnetic (TEM) signals are greatly affected by spike noise, resulting in poor electromagnetic data imaging effect. So, we simulated the theoretical TEM signal and develop a noise reduction method based on the transient characteristics analysis of theoretical signal and field noise. The method uses stationary wavelet transform (SWT) to analyze the multi-resolution data. First, the TEM data are decomposed into detailed and approximate coefficients. Then the logarithmic slope and threshold of the field data are calculated to identify the noise in the detail coefficient, and the corresponding detail coefficient is processed to reduce the noise. Finally, the pure TEM data is accurately reconstructed by stationary wavelet inversion (ISWT).