MD Loss: Efficient Training of 3-D Seismic Fault Segmentation Network Under Sparse Labels by Weakening Anomaly Annotation

Data-driven fault detection has been regarded as a 3-D image segmentation task. The models trained from synthetic data are difficult to generalize in some surveys. Recently, training 3-D fault segmentation using sparse manual 2-D slices is thought to yield promising results, but manual labeling has many false negative labels (FNLs) (abnormal annotations), which is detrimental to training and consequently to detection performance. Motivated to train 3-D fault segmentation networks under sparse 2-D labels while suppressing FNLs, we analyze the training process gradient and propose the mask dice (MD) loss. Moreover, the fault is an edge feature, and current encoder–decoder architectures widely used for fault detection (e.g., U-shape network) are not conducive to edge representation. Consequently, fault-net is proposed, which is designed for the characteristics of faults, employs high-resolution propagation features, and embeds multiscale compression fusion block to fuse multiscale information, which allows the edge information to be fully preserved during propagation and fusion, thus enabling advanced performance via few computational resources. The experiment demonstrates that MD loss supports the inclusion of human experience in training and suppresses FNLs therein, enabling baseline models to improve performance and generalize to more surveys. Fault-Net is capable of providing a more stable and reliable interpretation of faults, and it uses extremely low computational resources and inference is significantly faster than other models. Our method indicates optimal performance in comparison with several mainstream methods.