An optimized stacking method for common-image gathers based on iterative smooth dynamic time warping

Prestack depth migration is crucial to seismic imaging. However, its application to complex structures is impeded by the limitations of migration velocity analysis. The inaccuracies in the reconstructed velocity model lead to curved events in common-image gathers, which consequently reduce the precision of stacking. Dynamic time warping (DTW) addresses curved event correction by matching the similarity between a reference trace and other traces. However, significant amplitude differences between traces or low signal-to-noise ratios can lead to mismatch, resulting in waveform distortion and poor trace alignment. To solve this problem, this paper proposes a new method based on the iterative smooth dynamic time warping (ISDTW) algorithm. This method introduces smoothness constraints and an iterative strategy to optimize the warping path and thereby enhances the accuracy and consistency of trace alignment. Especially under complex geological conditions, ISDTW demonstrates improved imaging performance. Synthetic and field data tests validate the notable advantages of ISDTW in reducing local matching errors and improving the quality of the stacked data.