Rock thin section component identification method, device and medium based on machine learning
By applying a machine learning-based neural network model in petroleum exploration and development, the rock sheet images are segmented and identified, and the problem of low recognition efficiency of rock sheet components in the existing technology is solved, achieving a more efficient and accurate recognition effect.
CN113435456BActive Publication Date: 2025-04-01CHINA PETROLEUM & CHEMICAL CORP +1
1 Cites 0 Cited by
Patent Information
- Application Number
- CN202110653812.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2021-02-10
- Filing Date
- 2021-06-11
- Publication Date
- 2025-04-01
- Estimated Expiration
- 2041-06-11
AI Technical Summary
Technical Problem
During the oil exploration and development process, the component areas in the rock flakes are complex, the manual identification efficiency in the prior art is low, and there are uncontrollable influencing factors.
Method used
The neural network model based on machine learning is used to segment and identify the rock sheet images. By obtaining the image of the sheet to be tested and inputting the recognition model, outputting the segmentation area and recognition results, the geometric characteristics of the rock sheet samples are finally determined.
Benefits of technology
It improves the recognition efficiency and accuracy of components in rock flakes, and reduces the dependence of manual identification and the influence of uncontrollable factors.
✦ Generated by Eureka AI based on patent content.
Abstract
The present application is about a method, device and medium for identifying rock slice components based on machine learning, and relates to the field of petroleum geological exploration. The method includes: obtaining at least one image of a thin slice to be tested; inputting the image of the thin slice to be tested into a rock slice image recognition model, and outputting a segmented area and a recognition result corresponding to the image of the thin slice to be tested; and determining the geometric features of the thin slice sample to be tested based on the segmented area and the recognition result. In the process of determining the type of the rock sample, the image of the thin slice to be tested is input into the rock slice image recognition model, and the segmented area and the recognition result corresponding to the image of the thin slice to be tested are obtained, and finally the geometric features of the thin slice sample to be tested are determined. In the process of rock identification, by means of a neural network model based on machine learning, the efficiency of identifying components in rock slices is improved by segmenting and identifying each component area in the image of the thin slice to be tested.
Need to check novelty before this filing date? Find Prior Art