Seismic First Break Picking Through Swin Transformer Feature Extraction

Abstract

The application of neural networks to seismic first breaks (FBs) picking research has been developed for many years. Numerous multi-trace FBs picking methods based on convolutional neural networks have been proposed. Among them, the pickup method of semantic segmentation networks based on fully convolutional networks (FCNs) is proven to have stronger noise immunity. However, when in the data with drastic variations in the FBs between local adjacent traces, because of the feature extraction method of FCNs for convergence information around the data, the network output feature edge tends to smooth, this leads to a flattening of the FBs of the multi-trace pickup, which is not conducive to picking traces with drastic inter-trace FBs variation. Therefore, we use Transformer, a weight-transfer model that relies on the self-attention mechanism to calculate weights between input and output data, to extract FBs features. The 2D seismic data are flattened into a 1D sequence along the time dimension of the trace and input to the network, the spatial information of the FBs of the adjacent traces is considered without disrupting the time-series semantic information. The correlation between any element of the sequence and other elements is calculated to obtain the sequence feature weights. We use Swin Transformer as the backbone and combine the features of U-shaped networks to design an end-to-end FBs picking network – STUNet. The results show that STUNet has higher FBs picking accuracy than current FCNs, and is more effective in local adjacent traces where the FBs time variation is drastic.