An adaptive multi-scale seismic high-resolution processing method fusing time-frequency features

By constructing a Transformer-based encoder-decoder network architecture, and combining a frequency enhancement module with a dual attention mechanism of adaptive multi-scale partitioning, the limitations of existing high-resolution seismic processing methods are overcome, and efficient seismic resolution enhancement for complex geological structures is achieved.

CN121276597BActive Publication Date: 2026-05-26北京源澜科技有限公司
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
北京源澜科技有限公司
Filing Date
2025-09-30
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing high-resolution seismic processing methods rely on assumptions and experience, and are insufficient in time-frequency feature fusion, multi-scale modeling, and model generalization capabilities, resulting in limited improvement in seismic resolution.

Method used

A Transformer-based encoder-decoder network architecture is constructed, which combines a frequency enhancement module, adaptive multi-scale partitioning, and a dual attention mechanism. The model is trained using K-Means energy clustering and a hybrid loss function to achieve time-frequency feature fusion and multi-scale seismic data processing.

Benefits of technology

It enhances the model's transferability and generalization capabilities, effectively handles seismic data from different regions, reduces reliance on professional experience, and improves the ability to identify seismic resolution, especially revealing local details and global trends in complex geological structures.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121276597B_ABST
    Figure CN121276597B_ABST
Patent Text Reader

Abstract

This invention discloses an adaptive multi-scale high-resolution seismic processing method that integrates time-frequency features, belonging to the field of seismic exploration data processing technology. This method constructs a Transformer-based encoder-decoder network architecture, combines continuous wavelet transform to achieve time-frequency feature fusion, and employs an adaptive multi-scale partitioning strategy based on K-Means energy clustering and a dual attention mechanism to capture local and global dependencies in seismic data, ultimately improving the resolution of the seismic data. This invention addresses the over-reliance of traditional methods on geological assumptions and professional knowledge, and the shortcomings of existing deep learning methods in utilizing time-frequency features and multi-scale modeling. It possesses stronger generalization and transfer capabilities, making it suitable for processing seismic data with complex geological structures.
Need to check novelty before this filing date? Find Prior Art