A novel wavefield reconstruction method using sparse representation and dictionary learning for RTM

Abstract Reverse time migration (RTM) is a well-established imaging technique that uses the two-way wave equation to achieve high-resolution of complex subsurface media. However, when using RTM for reverse extrapolation, source wavefield needs be stored cross-correlation with backward wavefield. This requirement results in significant storage burden on computer memory. paper introduces reconstruction method combines sparse representation compress substantial amount crucial information The K-SVD algorithm train an adaptive dictionary, learned from training dataset consisting image patches. For each timestep, divided into patches, which are then transformed series coefficients trained dictionary via batch-orthogonal matching pursuit algorithm, known its accelerated coding process. novel essentially attempts transform domain reduce burden. We used several evaluation metrics explore impact parameters performance. conducted numerical experiments acoustic and compared two methods checkpointing techniques strategies our proposed method. Additionally, we extended application elastic RTM. tests demonstrate this can efficiently data, while considering both computational efficiency accuracy.