Seismic Shot-Record Arrival Detection With Energy Ratios

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional seismic data processing techniques for first arrival picking are time-consuming and prone to inaccuracies, with existing automated methods leading to false picks, and there is a need for improved automation and reduction in errors.

Innovation Solution

A method involving texture-based segmentation using fuzzy c-means clustering and energy ratio enhancement to identify first arrival picks in seismic data, utilizing descriptors like mean, second difference moment, energy, and contrast variance to segment seismic images and enhance cluster accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual picking is used for first arrival identification, then accuracy is maintained through human decision making, but time consumption increases significantly

Engineering Contradiction:

Improvefirst arrival picking accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent introduces texture segmentation as an intermediary process between raw seismic data and first arrival picking. By converting seismic data into texture images and segmenting them to highlight first arrival events, the system bridges the gap between automated processing and manual picking accuracy, enabling automated methods to achieve over 99% accuracy on synthetic data and over 80% on real data.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If conventional automated picking methods are used, then processing speed increases, but false picks occur due to inaccuracies

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidpicking accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent transforms the seismic data representation by changing parameters from raw amplitude data to texture features (energy, contrast, homogeneity, correlation). This parameter transformation enables automated algorithms to reliably identify first arrivals by analyzing texture patterns rather than raw waveforms, eliminating false picks while maintaining high processing efficiency.

Inventive Principle:

Principle #35Parameter changes

3Ease of manufacture

If conventional seismic data processing techniques are used, then existing methods are simple to implement, but they require significant human effort and are expensive

Engineering Contradiction:

Improveimplementation simplicityVSAvoidautomation level

Core Design Contradiction:

Ease of manufactureVSExtent of automation

Solution Approach 1:

The patent replaces manual mechanical picking operations with automated image processing and texture segmentation algorithms. By substituting human visual inspection and manual annotation with computer-based texture analysis, the system achieves full automation while maintaining ease of implementation through standard image processing techniques like fuzzy c-means clustering.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12405395B1Method and system for identifying geological formation and drilling hydrocarbon well

Publication Date: 2025.09.02 KING FAHD UNIVERSITY OF PETROLEUM AND MINERALS

AI summary

A method for first arrival picking of seismic data includes receiving a seismic shot record data set from a seismic event, extracting texture features from the seismic shot record data set, calculating an energy ratio matrix from the seismic shot record data set, using fuzzy c-means to cluster the texture features into a strong linear cluster, a moderate linear cluster, and a random noise cluster, calculating a first arrival class matrix from the moderate linear cluster, multiplying the first arrival class matrix by the energy ratio matrix to obtain a first arrival pick matrix, and identifying first arrival picks from the first arrival pick matrix.