Sedimentary Facies Image Segmentation Method Based on Neural Network
An image segmentation and neural network technology, which is applied in image analysis, image data processing, instruments, etc., can solve the problems of complex sedimentary facies images, high noise, complicated labels, etc. The method is convenient and efficient, the classification result is accurate, and the calculation amount is reduced Effect
Active Publication Date: 2021-05-04
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF7 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the ILBD algorithm is not suitable for the segmentation of sedimentary facies images because the sedimentary facies images are very complex, and the artificial input labels in the ILBD algorithm are very strict, and it is required to input labels for areas that are not connected to each other, regardless of whether these areas are of the same type
This requirement is unrealistic in sedimentary facies image segmentation, because sedimentary facies images are very complex, there are many disconnected areas, and manual input of each label is very cumbersome and noisy.
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0022] Embodiments of the present invention will be further described below in conjunction with the accompanying drawings.
[0023] Before the solution of the present invention is described in detail, the deep neural network algorithm and the SLIC image segmentation algorithm are described first.
[0024] Deep Neural Network (DNN) is currently the basis of many artificial intelligence applications. The outstanding performance of DNN stems from its ability to use statistical learning methods to extract high-level features from raw sensory data and obtain input space in a large amount of data. effective representation.
[0025] Neural networks take their inspiration from the concept that neurons involve computing weighted sums of input values. These weighted sums correspond to the scaling of the synaptic completion value and its combination with the neuron value. Also, because the computation is associated with the neuron level and it is a simple linear algebra operation, neur...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a method for segmenting sedimentary facies images based on a neural network, which combines human experience with the calculation speed of a computer, organically combines the work experience of geological workers with the calculation ability of a machine, that is, artificially input layers Label information, and the machine calculates the layer classification information based on these label information, and finally tracks the area boundary according to the layer classification information, so as to achieve human-computer interaction and make full use of it. Specifically, the superpixel algorithm SLIC is used for preliminary processing on the sedimentary facies map, and it is divided into homogeneous regions, and then the deep neural network is used to classify the image into a corresponding number of categories, which will separate the different types of sedimentary facies. The pixel areas of the algorithm are fused into a category area to achieve the purpose of reducing the amount of calculation. Adding seismic attribute data to the sample data makes the classification results more accurate, and placing the data on the grid structure is more convenient and efficient because the data is distributed in the spatial grid.
Description
technical field [0001] The invention belongs to the technical field of seismic exploration and relates to seismic data interpretation technology, in particular to a sedimentary facies image segmentation method. Background technique [0002] Seismic data interpretation is a very important link in seismic exploration, and sedimentary facies image interpretation is an important part of it. Sedimentary facies in geology reflect the characteristics and evolution process of the geographical environment in the geological period. Therefore, it is of great theoretical significance to study the sedimentary relative to understand the paleogeographical environment and the historical evolution of the crust in the geological era. It is of great practical significance to the design and planning of engineering construction. Nowadays, the analysis of sedimentary facies mainly relies on manual means to draw manually. The painters use their professional knowledge and experience to judge wheth...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/11
Inventor 姚兴苗颜博胡光岷
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA