Seismic horizon tracking based on the TransUnet model

ABSTRACT Seismic horizons are of paramount importance for building structure models and stratigraphic interpretations. However, currently, horizon tracking is generally obtained through manual tracking or a combination of traditional automatic tracking and manual tracking, which is a highly time-consuming and error-prone process. Despite the many methods developed for automated horizon tracking, tracking horizons with complex features still faces numerous challenges. We develop a method for horizon tracking using the TransUnet model, which can be applied to track individual horizons and multiple horizons simultaneously. This method is implemented through deep-learning semantic segmentation. Training is conducted using a deep-learning encoder-decoder structure to establish the mapping relationship between seismic data and horizon data. The TransUnet model builds upon the U-net model by incorporating transformer modules, enabling the model to possess local and global attention capabilities. The ultimate results of horizon tracking reveal that the TransUnet model is more proficient at tracking horizons in areas with complex seismic reflections than the U-net model. In single and multiple horizon tracking, the root mean square error and the coefficient of determination (R2) calculated from horizons predicted by the TransUNet model and manually tracked horizons are significantly lower than those of the U-net model. This research determines that our method is more efficient and accurate in tracking the horizons of 3D seismic volumes compared with traditional approaches.