Reservoir fluorescent slice oiliness and maturity evaluation method based on artificial intelligence

Through artificial intelligence-based methods, fluorescent image data is used for preprocessing and feature extraction, combined with the 1D convolutional neural network model, the accuracy, consistency and efficiency of fluorescence image analysis in the existing technology is solved, achieving high efficiency and accuracy Evaluation of oil content and maturity of reservoir fluorescent flakes.

CN119477828AActive Publication Date: 2025-02-18YANGTZE UNIVERSITY
View PDF 5 Cites -1 Cited by

Patent Information

Application Number
CN202411502881.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-25
Publication Date
2025-02-18
Estimated Expiration
2044-10-25

AI Technical Summary

Technical Problem

The existing fluorescence image analysis methods have shortcomings in accuracy, consistency and efficiency, which are difficult to meet the needs of large-scale oil and gas exploration, and relying on professionals leads to high costs.

Method used

Using an artificial intelligence-based method, we use collect fluorescent image data, preprocess images, extract advanced features, fuse multi-dimensional features, and build a 1D convolutional neural network model to predict maturity and oil content.

Benefits of technology

It improves the accuracy and consistency of analysis results, significantly improves processing efficiency, reduces dependence on professionals, and greatly reduces project costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119477828A_ABST
    Figure CN119477828A_ABST
Patent Text Reader

Abstract

The invention provides a reservoir fluorescent slice oiliness and maturity evaluation method based on artificial intelligence, and the method comprises the steps: firstly collecting oil-containing fluorescence image data and corresponding RO values to construct an image library, then carrying out the data preprocessing, extracting features through a convolutional neural network, carrying out the clustering analysis, carrying out the visualization, finding out an abnormal image, carrying out the standardization processing, and storing, errors caused by different environments and devices are reduced as much as possible. And then oil-containing region division is carried out, extraction is carried out according to a threshold value after color space conversion, mask processing and shape filtering are carried out to remove oil-immersed minerals, and regions are extracted and an image is saved after edge detection. And performing multi-dimensional feature fusion extraction, including extraction and processing of color, gray scale and texture features. Then, a 1D convolutional neural network model is built, features are input, and parameters are optimized; and finally, constructing a model to predict maturity and oiliness, and inputting the model after dimension adjustment. The method combines deep learning and computer vision, and has the advantages of high efficiency, accuracy, multi-dimensional fusion and the like.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Method for rapidly analyzing rock debris fluorescence image

    CN103927547A

  • Visible light image-based walnut maturity detection and prediction method

    CN113192026A

  • No-reference color image quality evaluation method based on double convolutional neural networks

    CN118314075A

  • Deposition apparatus equipped with a shutter for opening and closing a linear source

    KR1020250111524A

  • System and Method for Detection of Carbonate Core Features from Core Images

    US20220067420A1

Cited By

  • Intelligent design method of concrete faced rockfill dam profile based on data feature fusion

    CN119918427A

  • Deep learning quantitative characterization method and system for oiliness of CT (Computed Tomography) scanning oil layer

    CN120318439A

  • Wheat gluten production quality evaluation method based on machine learning

    CN120411955A

  • Wind power blade tip detection method and system based on brightness adaptation and multi-scale attention

    CN121564106A

  • A wind turbine blade tip detection method and system based on brightness self-adaption and multi-scale attention

    CN121564106B