Abstract:Although the excitation-time imaging condition offers a lower memory consumption and higher computational efficiency compared to cross-correlation imaging condition, it has not been widely used in industrial applications because of the accuracy problem of traveltime calculation and the influence of low-wave-number noise. In this paper, we introduce the physics-informed neural network (PINN) algorithm to achieve a high-precision traveltime calculation of the source forward wavefield. Subsequently, we introduce a technique for high-precision wavefield decomposition of the reverse-time wavefield via the optical flow vector, enabling us to realize a correlation-weighted stacking imaging of each wavefield. Model experiments and real data processing show that the proposed traveltime calculation algorithm based on PINN offers high accuracy and good applicability in the excitation time reverse-time migration imaging of complex models, and correlation-weighted stacking imaging based on optical flow vector-based wavefield separation can significantly suppress the noise with low wavenumber and achieve high-precision imaging of complex models.

What problem does this paper attempt to address?

The problems that this paper attempts to solve mainly focus on two aspects: 1. **Accuracy problem of travel - time calculation**: In reverse - time migration (RTM) based on excitation - time imaging conditions (ETIC), accurate travel - time calculation is crucial for improving imaging accuracy. However, existing methods such as solving hyperbolic equations have limitations in calculation accuracy and computational efficiency. This paper introduces an algorithm based on physics - informed neural networks (PINN) to achieve high - accuracy travel - time calculation to overcome these problems. 2. **Suppression problem of low - wavenumber noise**: In the traditional RTM imaging process, due to the failure to effectively separate reflected waves, a large amount of low - wavenumber noise is generated, which seriously affects the authenticity and accuracy of imaging. This paper proposes a wave - field separation technique based on optical - flow vectors (OF). Through correlation - weighted superposition imaging, the low - wavenumber noise is significantly suppressed and the imaging accuracy of complex models is improved. ### Specific solutions 1. **Travel - time calculation based on PINN**: - Use the PINN algorithm to solve the two - dimensional hyperbolic equation to achieve high - accuracy travel - time calculation. - Solve the source - point singularity problem through the hyperbolic equation in factorized form, improving the calculation accuracy of travel - time. - Use the fully - connected neural network (FCNN) to estimate the unknown travel - time factor and train the network by optimizing the loss function. 2. **Wave - field separation based on optical - flow vectors**: - Introduce optical - flow vectors (OF) to separate the received wave - field, avoiding the instability of the traditional Poynting vector in wave - field separation. - Improve the accuracy of wave - field separation through the iterative process and global smoothing constraints. - Correlation - weighted superposition imaging of the wave - fields in four separated directions effectively suppresses low - wavenumber noise and improves the imaging quality. ### Experimental verification - Through model experiments and actual data processing, the high - accuracy and good applicability of the travel - time calculation algorithm based on PINN in excitation - time reverse - time migration imaging of complex models are verified. - The wave - field separation technique based on optical - flow vectors significantly suppresses low - wavenumber noise and achieves high - accuracy imaging of complex models. In conclusion, by introducing PINN and optical - flow vector techniques, this paper solves the problems of travel - time calculation accuracy and low - wavenumber noise suppression in RTM imaging, providing new methods and technical support for high - accuracy imaging of complex geological structures.

Excitation time imaging condition reverse time migration based on physics-informed neural network traveltime calculation with wavefield decomposition using optical flow vector

A Model Based Algorithm for Fast DPIV Computing

An efficient least-squares reverse time migration in image domain

Elastic Reverse Time Migration Without Wavefield Decomposition

Isotropic Elastic Reverse Time Migration

Reflected wave least squares reverse time migration with angle illumination compensation

Forward Imaging Neural Network with Correction of Positional Misalignment for Fourier Ptychographic Microscopy.

A Fast and Efficient DPIV Algorithm Using IAFM Models

PINNPStomo: Simultaneous P- and S-wave seismic traveltime tomography using physics-informed neural networks with a new factored eikonal equation

Wave simulation in non-smooth media by PINN with quadratic neural network and PML condition

Reverse Time Migration Based on Normalized Wavefield Decomposition Imaging Condition

Seismic traveltime simulation for variable velocity models using physics-informed Fourier neural operator

Dynamic-excitation-based Steady-State Non-Line-of-sight Imaging Via Multi-Branch Convolutional Neural Network

Physics-informed Neural Networks for 3D Seismic Travel Time Tomography

Stable neural network-based traveltime tomography using hard-constrained measurements

Dense velocity reconstruction from particle image velocimetry/particle tracking velocimetry using a physics-informed neural network

Deep Learning for Enhancing Multi-Source Reverse Time Migration

Time-lapse velocity imaging via deep learning

Optical flow inversion for remote sensing image dense registration and sensor's attitude motion high-accurate measurement

Modeling multisource multifrequency acoustic wavefields by a multiscale Fourier feature physics-informed neural network with adaptive activation functions