Application of Gradient Learning Scheme to Pixel-Based Inversion for Transient EM Data

Guo, Rui; Li, Maokun; Yang, Fan; Xu, Shenheng; Fang, Guangyou; Abubakar, Aria

doi:10.1109/compem.2018.8496518

Cited by 8 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of descent direction searching, Yin and Hodges [52] proposed a method using the simulated quenching method. Guo et al [53] proposed an inversion method based on the supervised descent method, which is a method to select the inverse descent direction by machine learning method. In addition, Zhu et al [54] proposed an artificial neural network (ANN) inversion method based on the principal components of the decay curve.…”

Section: Inversionmentioning

confidence: 99%

The Progress of the Helicopter-Borne Transient Electromagnetic Method and Technology in China

Xue

2020

IEEE Access

View full text Add to dashboard Cite

The central and western regions of China have appropriate conditions for mining development, but there are many deserts, swamps and forest-covered areas in this vast land. As a result, the traditional on-ground methods are difficult to carry out efficiently in these areas. The airborne transient electromagnetic method can implement detection over large areas rapidly and efficiently, and play an important role in terms of resource and environment detection. In recent years, two Helicopter-borne Transient electromagnetic (HTEM) systems, CHTEM-I and CAS-HTEM, were developed in China. In this paper, the difficult issues in the development of HTEM hardware and software will be analyzed, firstly. By tracing the specific technical paths to solve these difficult issues, the latest research achievements in the development process of CHTEM-I and CAS-HTEM will be expounded. Finally, we would like to briefly discuss the research and development direction of HTEM technology in China to promote further development of the research and application of this technology.

show abstract

Section: Inversionmentioning

confidence: 99%

The Progress of the Helicopter-Borne Transient Electromagnetic Method and Technology in China

Xue

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In recent years, the supervised descent method (SDM) (Xiong and De la Torre 2013) has been applied for 2D microwave imaging, which incorporates prior information in a flexible manner and shows a good ability of generalization (Guo et al 2018. The SDM is developed based on the Gauss-Newton (GN) method, where a learning technique is applied to compute the descent directions rather than the derivation of cost function in the GN method.…”

Section: Introductionmentioning

confidence: 99%

Neural network-based supervised descent method for 2D electrical impedance tomography

et al. 2020

View full text Add to dashboard Cite

Objective: In this work, we study the application of the neural network-based supervised descent method (NN-SDM) for 2D electrical impedance tomography. Approach: The NN-SDM contains two stages: offline training and online prediction. In the offline stage, neural networks are iteratively applied to learn a sequence of descent directions for minimizing the objective function, where the training data set is generated in advance according to prior information or historical data. In the online stage, the trained neural networks are directly used to predict the descent directions. Main results: Numerical and experimental results are reported to assess the efficiency and accuracy of the NN-SDM for both model-based and pixel-based inversions. In addition, the performance of the NN-SDM is compared with the linear SDM (LSDM), an end-to-end neural network (E2E-NN) and the Gauss–Newton (GN) method. The results demonstrate that the NN-SDM achieves faster convergence than the LSDM and GN method, and achieves a stronger generalization ability than the E2E-NN. Significance: The NN-SDM combines the strong non-linear fitting ability of the neural network and good generalization capability of the supervised descent method (SDM), which also provides good flexibility to incorporate prior information and accelerates the convergence of iteration.

show abstract

“…In the TEM survey, a primary electromagnetic (EM) impulse is transmitted to the underground and then a secondary EM field induced by the eddy current can be measured. These secondary field data contain abundant geoelectrical information and we can transform these data into visualized underground structure by means of inversion method [3]- [6].…”

Section: Introductionmentioning

confidence: 99%

A Simple and Robust Total Variation-Based Inversion Scheme for Transient Electromagnetic Data

Rao

Gao

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Transient electromagnetic (TEM) data are conveniently inverted to visualized underground structure with the widely used Occam's inversion. However, in sedimentary environments, Occam's inversion performs poorly in reproducing the sharp boundaries as it produces smoothed results of subsurface geophysical properties. Here, we develop a simple and robust total variation (TV) based inversion scheme for TEM data to preserve sharp boundaries and improve the accuracy of recovered underground structure. To solve the standard L2-TV optimization problem, we propose a reconstructing minimization (RM) approach. The TV-based inversion scheme is divided into two phases. In Phase I the misfit is brought down to a desired level utilizing Occam's inversion. In Phase II we utilize RM approach to implement TV regularization until the convergent result is achieved. Taking advantage of the excellent stability of Occam's inversion, the TVbased inversion scheme has the characteristics of robustness to different initial models. The switch process between Occam's inversion and the RM approach is rather simple to realize due to their similar solution forms. The synthetic and field data examples validate the efficiency and accuracy of the proposed TV-based inversion scheme. We further present a novel analysis based on reconstruction of stabilizer to illustrate the reason for their distinctive behaviors of TV-based inversion and Occam's inversion.INDEX TERMS Reconstructing minimization approach, sharp boundaries, total variation, transient electromagnetic inversion.

show abstract

Application of Gradient Learning Scheme to Pixel-Based Inversion for Transient EM Data

Cited by 8 publications

References 13 publications

The Progress of the Helicopter-Borne Transient Electromagnetic Method and Technology in China

The Progress of the Helicopter-Borne Transient Electromagnetic Method and Technology in China

Neural network-based supervised descent method for 2D electrical impedance tomography

A Simple and Robust Total Variation-Based Inversion Scheme for Transient Electromagnetic Data

Contact Info

Product

Resources

About