Deep-learning methods have proved successful recently for solving problems in image analysis and natural language processing. One of these methods, convolutional neural networks (CNNs), is revolutionizing the field of image analysis and pushing the state of the art. CNNs consist of layers of convolutions with trainable filters. The input to the network is the raw image or seismic amplitudes, removing the need for feature/attribute engineering. During the training phase, the filter coefficients are found by iterative optimization. The network thereby learns how to compute good attributes to solve the given classification task. However, CNNs require large amounts of training data and must be carefully designed and trained to perform well. We look into the intuition behind this method and discuss considerations that must be made in order to make the method reliable. In particular, we discuss how deep learning can be used for automated seismic interpretation. As an example, we show how a CNN can be used for automatic interpretation of salt bodies.
We consider the problem of estimating subsurface quantities such as velocity or reflectivity from seismic measurements. Because of a limited aperture and band‐limited signals, the output from a seismic prestack reconstruction method is a distorted or blurred image. This distortion can be computed using the concept of resolution function, which is a quantity readily accessible in the Fourier space of the model. The key parameter is the scattering wavenumber, which at a particular image point is defined by the incident and scattered ray directions in a given background model. Any location in any background model can be considered. In general, the resolution function will depend on the following four quantities: the background velocity model, the frequency bandwidth, the wavefield type and the acquisition geometry.
We first establish the resolution function for a general scattering model assuming local reaction. We then adapt this result for two well‐known scattering models: Born and Kirchhoff. For each of these approximations the corresponding resolution function is derived and discussed. Finally, by employing a simple synthetic data example we demonstrate the ability of the resolution function to predict the image distortions.
A B S T R A C TAdaptations of existing triaxial cells for ultrasonic P-and S-wave measurements are well documented. This paper proposes further modification of such a cell so that also resistivity measurements can be carried out simultaneously at reservoir conditions. By employing the top cap and the pedestal of the cell as electrodes, axial resistivity measurements are now feasible. In order to minimize the polarization effect of this two-electrode arrangement, careful analyses have been carried out to optimize the choice of electrode coating and measurement frequency band. Radial resistivity measurements are also included in the system by introducing a strap-electrode system.In a reservoir under production changes in both saturations, temperature (if steam injection) and stresses can take place. Therefore the modified triaxial system should be able to measure the integrated effects on the acoustic parameters and electric responses caused by variations in each of these parameters. The feasibility of the system to obtain such reliable information is demonstrated, employing a small selection of core samples. In the future such combined measurements on reservoir core samples can be used to link both seismic and electromagnetic observations to the actual earth model and constrain both modelling and inversion.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.