A denoising method based on wavelet analysis is presented for the removal of noise (background noise and random spike) from time domain electromagnetic (TEM) data. This method includes two signal processing technologies: wavelet threshold method and stationary wavelet transform. First, wavelet threshold method is used for the removal of background noise from TEM data. Then, the data are divided into a series of details and approximations by using stationary wavelet transform. The random spike in details is identified by zero reference data and adaptive energy detector. Next, the corresponding details are processed to suppress the random spike. The denoised TEM data are reconstructed via inverse stationary wavelet transform using the processed details at each level and the approximations at the highest level. The proposed method has been verified using a synthetic TEM data, the signal‐to‐noise ratio of synthetic TEM data is increased from 10.97 dB to 24.37 dB at last. This method is also applied to the noise suppression of the field data which were collected at Hengsha island, China. The section image results shown that the noise is suppressed effectively and the resolution of the deep anomaly is obviously improved.

The small-loop transient electromagnetic (TEM) method has made great contributions in urban underground space detection in recent years due to its small working space. For the problems of single sampling rate and low sampling rate of traditional receivers, a dual sampling rate receiver is designed to record electromagnetic response with high accuracy. In addition, this paper employs the passive clamping technique and oscillation absorption technique to realize the falling edge linear fast shutdown of the triangle and trapezoidal wave emission currents, and to improve the excitation capability of the shallow target. In order to avoid the saturation of the early response, the primary field intensity of the center point is reduced by bucking coil. In order to solve the problem of high precision data interpretation in shallow layer, this paper adopts the method of remaining primary field removal and response splicing to splice the pure secondary field response of triangle wave and trapezoidal wave current waveform, and realizes the joint resistivity imaging of electromagnetic data. According to laboratory test results, the transmitter can output triangle wave and trapezoidal wave with adjustable parameters; the sampling rate of the receiver is 64 kHz and 256 kHz, which can effectively record the electromagnetic response and current waveform. The field experimental results exhibit that the small-loop electromagnetic detection system can effectively measure the shallow anomalies in urban underground space, which demonstrates the effectiveness of the dual waveform combined small-loop detection system for improving the accuracy of shallow exploration and promoting the development of geophysical refinement detection process.

The induced polarization effects magnetic field measurement method based on the superconductive quantum interfere device (SQUID) and transient electromagnetic (TEM) system can improve the exploration accuracy of metal minerals. As the core component of the SQUID-TEM system based on magnetic source, the current transmitter should reach consensus in three key aspects: high power, controllable turn-off time and combined current waveforms. In order to solve the problem of single transmitting current waveform and uncontrollable turn-off time in traditional current transmitter, a topology structure of current transmitter with multiple waveform combinations in time domain is proposed in this paper. By using the passive multi-group clamp technology and RLC series resonance technology, the falling edge turn-off time of trapezoidal and triangular transmitting current can be controlled, and the pulse width of semi-sinusoidal transmitting current can be adjusted. The laboratory and field experiments are implemented to test the performance of the current transmitter. Compared with the commercial PROTEM67 transmitter, the test results show that it can successfully transmit three current waveforms and their combination waveforms, the current amplitude is more than 100A, and the transmitting current turn-off time is controllable. The field measurement response results show that the combined multi-waveform transmit system can effectively measure the induction field and the polarization field generated by the Cole-Cole polarization ring, which provides technical support for the high-precision detection of induction-polarization effects based on the SQUID-TEM system.

The small-loop time-domain electromagnetic (TDEM) method has been widely used in urban underground space detection in recent years due to the small workspace requirements. Due to the small side length of the transmitting coil of the small-loop electromagnetic method, more difficulties have been introduced in modeling and instrument development. The traditional modeling method cannot include source calculation, and the error becomes significantly large when calculating the initial field of the small-loop, and the cross iteration of electric and magnetic fields also increases the modeling error of the small-loop. Therefore, a high-precision 3-D small-loop source TDEM modeling method need to be proposed to provide the theoretical basis for feature analysis, inversion, and instrument parameter design. In this article, the electromagnetic wave equations are adopted as the controlling equations and discretized based on the Crank–Nicolson finite-difference time-domain (CN-FDTD) method. The entire computational space, including air and ground, is subdivided into sources to support 3-D small-loop TDEM modeling for shallow anomalous body conditions. Furthermore, the iterative formulas of the electromagnetic wave equations in the complex frequency-shifted perfect match layer (CFS-PML) are derived, the reflection errors are largely suppressed, and the modeling accuracy is significantly improved. Finally, the effectiveness of the improved method is verified by homogeneous models, layered models, and complex anomaly models. In addition, analyzing the propagation characteristics of the small-loop can guide the setting of the receiver sampling rate and improve the accuracy of detection. The results show that the improved method can achieve stable, low-memory, and high-precision 3-D small-loop source TDEM modeling, which can provide theoretical support for the application of the TDEM method in urban and shallow detection.

With the development of electronic technology, time-domain electromagnetic (TDEM) transmitters are now capable of emitting various waveforms such as trapezoidal, triangular, and semi-sinusoidal waveforms, as well as their combinations. However, the application of multi-waveform combination detection methods is still immature, and modeling methods are required to analyze the characteristics of different waveforms and complete data interpretation. Based on the Crank-Nicolson finite difference time-domain (CN-FDTD) method for discretizing the electromagnetic wave equations, the current waveform can be converted into current surface density for source-included calculations, enabling high-precision modeling of the responses to different current waveforms. The responses to multiple transmission current waveforms were compared and analyzed using various models, and the results showed that selecting targeted excitation parameters based on the characteristics of the target can improve detection sensitivity. We applied the multi-waveform combination detection method to the actual exploration in the field, obtained the pure secondary field response by subtracting the primary field of the actual current waveform from the measured data, converted the responses of different transmission current waveforms into step wave responses by deconvolution method, and performed resistivity imaging. The results of modeling and data interpretation demonstrate the effectiveness of the multi-waveform combination detection method in improving geological exploration accuracy, and promoted the development of fine-scale geophysical exploration.

To address the issue of low detection accuracy caused by the similarity between inspection targets and backgrounds in power transmission line inspection images, this paper designs a visual defect inspection framework for electrical components based on weakly supervised contrastive learning pretraining. First, two power transmission line inspection datasets are constructed: one containing coarse labels and another with fine-grained labels. The coarse label dataset is used for weakly supervised contrastive learning pretraining, where the coarse label information distinguishing the target from the background is utilized to perform feature metric learning on one-dimensional (global) and two-dimensional (local) features. This significantly improves the ability to distinguish between background and target features in local regions. The fine-grained label dataset is used for supervised target detection model transfer learning, which provides precise target location and category information, further enhancing the model’s accuracy and robustness in detecting targets against complex backgrounds. Additionally, this paper introduces a deformable convolution operation at the global image level to learn the boundaries between the background and the target. Furthermore, a novel multi-path coordinated attention mechanism is designed to dynamically optimize the boundary offset, further enhancing the global feature distinction and target detection performance. Experimental results show that the proposed method improves the detection accuracy from 76.0% to 79.4% compared to the original algorithm. When transferred to similar target detection algorithms, the method achieves an improvement ranging from 0.7% to 2.8%, effectively enhancing the performance of existing algorithms in detecting targets in complex background environments. The dataset and code are available at https://github.com/19971483525/IEEE-TIM.