Accurate interpolation of seismic data is crucial for improving the quality of imaging and interpretation. In recent years, deep learning models such as U-Net and generative adversarial networks have been widely applied to seismic data interpolation. However, they often underperform when the training and test missing patterns do not match. To alleviate this issue, here we propose a novel framework that is built upon the multi-modal adaptable diffusion models. In the training phase, following the common wisdom, we use the denoising diffusion probabilistic model with a cosine noise schedule. This cosine global noise configuration improves the use of seismic data by reducing the involvement of excessive noise stages. In the inference phase, we introduce the denoising diffusion implicit model to reduce the number of sampling steps. Different from the conventional unconditional generation, we incorporate the known trace information into each reverse sampling step for achieving conditional interpolation. To enhance the coherence and continuity between the revealed traces and the missing traces, we further propose two strategies, including successive coherence correction and resampling. Coherence correction penalizes the mismatches in the revealed traces, while resampling conducts cyclic interpolation between adjacent reverse steps. Extensive experiments on synthetic and field seismic data validate our model’s superiority and demonstrate its generalization capability to various missing patterns and different noise levels with just one training session. In addition, uncertainty quantification and ablation studies are also investigated.

Seismic data are generally sampled irregularly and sparsely along spatial coordinates because economic costs and obstacles hinder the regular arrangement of geophones in the field. Thus, the sampled seismic data often contain missing traces which result in difficulties for later processing steps. To alleviate this issue, versatile interpolation methods have been developed to interpolate the missing traces. However, the existing models for recovering seismic data with consecutively missing traces in a large amplitude range tend to produce artifacts and blurred signal details. We propose in this paper a hybrid loss guided coarse-to-fine model which consists of a coarse network and a refinement network to allow different regions of seismic data to be recovered in different stages. The coarse network is designed to reconstruct the strong signals and the refinement network is implemented subsequently to recover the weak signals. In addition, the refinement network focuses its attention on the areas which are not well recovered by the coarse network via a weight-masked mechanism. By resorting to the hybrid loss function L1+SSIM+Relativistic Average Least-Square Generative Adversarial Network (RaLSGAN), our model enables more accurate and realistic signal details to be reconstructed. Experiments with synthetic and field data demonstrate that our model is superior to the existing mainstream approaches and the role of the key components is also investigated through ablation studies.

Picking the first arrival of prestack gathers is an indispensable step in seismic data processing. To enhance the efficiency of seismic data processing, some deep-learning-based methods for first-arrival picking have been developed. However, when applying currently trained models to data that significantly differ from the training set, the results are often suboptimal. We refer to this predictive scenario as cross-survey picking. Therefore, further improving model generalization for accurate cross-survey picking has become an urgent problem. To overcome the problem, we develop a multistage picking method called multistage segmentation picking network (MSSPN), which breaks down the complex picking task into four stages. In the first stage, we develop a coarse segmentation network to recognize a rough trend of first arrivals. Second, a robust trend estimation method is developed in the second stage to further obtain a tighter range of first arrivals. Third, a refined segmentation network is conducted in the third stage to pick high-precision first arrivals. Finally, we develop a velocity constraint-based postprocessing strategy to remove the outliers of network pickings. Extensive experiments indicate that MSSPN outperforms current state-of-the-art methods under the cross-survey test situation in terms of the metrics of accuracy and stability. Particularly, MSSPN achieves 94.64% and 89.74% accuracy under the cross-survey field cases of the median and low signal-to-noise ratio data, respectively.

In seismic exploration, first break (FB) picking is a crucial aspect in determining subsurface velocity models, significantly influencing the placement of wells. Many deep neural networks (DNNs)-based automatic picking methods have been proposed to accelerate this process. Significantly, the segmentation-based DNN methods provide a segmentation map and then estimate FB from the map using a thresholding technique. However, these automatic methods applied in field datasets cannot ensure robustness, especially in the case of a low signal-to-noise ratio (SNR). In this paper, we introduce uncertainty quantification into FB picking and propose a novel uncertainty-based picking deep learning network called UPNet. UPNet specifically consists of two DNNs. A Bayesian network infers a posterior distribution by sampling the segmentation map of FB. Subsequently, a regression network integrates the segmentation map, the original trace, and the low-frequency trace to infer the FB trace by trace. Finally, a decision-making method provides the final FB based on uncertainty analysis, offering people robust FB. Importantly, UPNet avoids post-processing to obtain FB using the threshold method, as in the segmentation-based picking methods, and instead provides the FB of each trace end-to-end. Moreover, UPNet not only estimates the uncertainty of the network output but can also filter out predictions with low confidence. Many experiments have shown that UPNet demonstrates higher accuracy and robustness than the deterministic DNN-based model, achieving state-of-the-art (SOTA) performance in field surveys. In addition, we verify that the calculated uncertainty is significant, which can serve as a reference for human decision-making.

In seismic exploration, identifying the first break (FB) is critical in establishing subsurface velocity models. Various automatic picking techniques based on deep learning have been developed to expedite this procedure. The most popular method involves using semantic segmentation networks to pick on a shot gather known as 2-dimensional (2-D) picking. Concretely, segmentation-based methods input a gather, and produce a binary segmentation map, where the highest value in each column represents the FB. However, currently designed segmentation networks make it difficult to ensure the horizontal continuity of the segmentation. FB jumps also exist in some areas, and it is challenging for current models to detect such jumps. Therefore, it is important to pick as much as possible and ensure horizontal continuity. To address this issue, we propose a novel network for 2-D seismic FB picking. We introduce the dynamic snake convolution (DSConv) from computer vision into U-Net and refer to the new model as Dynamic-Snake U-Net (DSU-Net). Specifically, we develop the original DSConv and propose a novel DSConv module, which can extract the horizontally continuous texture in the shallow features of the shot gather. Many experiments show that DSU-Net demonstrates higher accuracy and robustness than the other 2-D segmentation-based models, achieving state-of-the-art (SOTA) performance in 2-D seismic field surveys. Particularly, it can effectively detect FB jumps and better ensure the horizontal continuity of FBs. Additionally, the ablation experiment and the anti-noise experiment, respectively, verify the optimal structure of the DSConv module and the robustness of the picking.

Velocity picking, a critical step in seismic data processing, has been studied for decades. Although manual picking can produce accurate normal moveout (NMO) velocities from the velocity spectra of prestack gathers, it is time-consuming and becomes infeasible with the emergence of a large amount of seismic data. Numerous automatic velocity picking methods have thus been developed. In recent years, deep learning (DL) methods have produced good results on the seismic data with medium and high signal-to-noise ratios (SNRs). Unfortunately, it still lacks a picking method to automatically generate accurate velocities in situations of low SNR. In this article, we propose a multiinformation fusion network (MIFN) to estimate stacking velocity from the fusion information of velocity spectra and stack gather segments (SGSs). In particular, we transform the velocity picking problem into a semantic segmentation problem based on the velocity spectrum images. Meanwhile, the information provided by SGS is used as a prior in the network to assist segmentation. The experimental results on two field datasets show that the picking results of MIFN are stable and accurate for the scenarios with medium and high SNR, and it also performs well in low SNR scenarios. Code is made publicly available at https://github.com/newbee-ML/MIFN-Velocity-Picking