Magnetic resonance sounding (MRS) measurements commonly suffer from a notably low signal-to-noise ratio (SNR). In recent years, many denoising methods have been developed, which have demonstrated the useful capability to improve the SNR. However, when MRS measurements are implemented on the sites with high noise levels, i.e., the human living environment, the conventional methods are helpless to recover the effective MRS signal, which is submerged in extensive environmental noise. In addition, the conventional methods that depend on signal models and the corresponding prior assumptions commonly rely on manual experience, which brings obstacles to the automation and efficiency of signal processing. To solve the above problems, we attempt to apply an intelligent denoising framework with a novel neural network as the basic tool for deep learning, called Dn-ResUnet in this article. The network extracts the features of the MRS signal through the encoder and decoder layers of the Dn-ResUnet structure, in which residual learning is adopted to accelerate the training process and improve denoising performance. Once the training is completed, the deep learning realizes adaptive denoising with no need for 1) prior assumptions of the MRS signal and noise; 2) optimal filter parameter tuning; or 3) expensive time cost. The comparison experiments demonstrate that the Dn-ResUnet model provides superior noise cancellation performance, especially it can replace the conventional methods to recover the effective MRS signals in noise levels down to an SNR of −30 dB. In addition, the noise attenuation tests are performed on synthetic and real data. The results show that the framework of deep learning yields a convincing performance in MRS signal denoising.

Nuclear magnetic resonance (NMR) is an important geophysical technique for the structure and properties measurement of porous media. The small-diameter NMR logging technology has been emerged as a useful tool in terms of evaluating the shallow surface hydrogeological parameters. However, the NMR logging is always suffering from the low signal-to-noise radio (SNR). This paper proposes a denoising network for NMR logging signal based on the combination of the convolutional neural network (CNN) and residual learning, called as Dn-ResNet. Firstly, a large amount training datasets are constructed according to the characteristics of the NMR logging signal and noise, which is used to train the network. Secondly, the useful features of the NMR logging signals are learned by the network during training processing, in which the residual learning mechanism is employed to improve the training efficiency and the denoising performance. As a result, the Dn-ResNet may realize adaptive denoising without manual filter parameters tuning, prior knowledge of NMR logging signals and preprocessing of the original noisy signals. The performance of the proposed denoising network is demonstrated on both the synthetic NMR logging signals and field data. The effective signals are obtained from the noisy signals at different SNRs. In addition, an attempt to use the well-trained Dn-ResNet to deal with a single record provides similar results compared to 200 stacks. It shows that the number of stacks may be reduced to shorten the measurement time and improve the measurement efficiency. The results show the effectiveness and robustness of the proposed approach, which enables the technology of deep learning to be developed in application of NMR logging data processing.

The magnetic resonance sounding (MRS) signal typically suffers from low signal-to-noise ratio (SNR). The MRS signal is severely distorted by noise, primarily harmonic and spiky noise. In terms of despiking, wavelet thresholding (WT) reconstructs the distorted content of the MRS signal, following isolation and elimination of the spiky sequence. However, WT cannot restore the MRS signal content completely when a series of spikes occurs within a given period of time. To solve this problem, a combined method of empirical mode decomposition (EMD) and WT for the removal of a series of spikes is proposed. EMD is first applied to decompose the noisy signal into several different components. The spikes that occur within a period of time are separated, the components without spikes are retained, and the components containing spiky events are selected and further processed by WT. After successively computing the wavelet coefficients of the selected components, the coefficients related to the spikes are isolated by threshold processing, and the subsequent wavelet reconstruction yields the sequence with the spikes removed. Finally, the denoised signal is obtained by adding the processed and retained components. The simulations on synthetic signals corrupted by artificial and real noise show that the proposed method improves the SNR with an accompanying improvement in the retrieval of the signal parameters. Moreover, the comparison results of the proposed and the WT methods suggest that the combined method efficiently removes a series of spikes.

Semi-airborne Electromagnetic Method (SAEM) is an emerging geophysical exploration method, which can be efficiently carried out in complex terrain conditions.SAEM signals are often mixed with various noise interference, which seriously affects the signal quality. However, the amount of SAEM data is extremely large, and the number of signals of a single survey line is in the millions, and the existing noise elimination methods usually intercept the signals in segments and then process them, which is inefficient and has strong subjective factors. To address the above problems, this paper proposes a recurrent self-coding neural network (RSCN) for one-stop noise cancellation of the whole measurement line signal. The method adopts the Gated Recurrent Unit(GRU) to obtain the full waveform state, and adopts the data segmentation technique to obtain the small-scale data and perform the convolutional coding. The state information as well as the coded data is used to extract important features using an attention mechanism and decoded. The self-encoding structure is updated cyclically to realize one-stop noise cancellation for the whole measurement line data. The test results of simulated and measured data show that the method proposed in this paper has a good denoising effect on the full survey line SAEM data, and is more adaptable to the SAEM data than the existing noise reduction methods.