A rapid identification method for organic pores and fractures in shale MAPs based on FCM

Abstract

This invention provides a rapid identification method for organic pores and fractures in shale MAPs based on FCM (Flexible Curve Analyzer), relating to the fields of oil and gas development and image segmentation. Specifically, it includes the following steps: preprocessing each image block using a self-guided filtering algorithm, and then combining the filtered image blocks into complete MAPs; segmenting the filtered MAPs based on the FCM algorithm; determining the pore type using the neighborhood gray-scale mean of the pores; and differentiating pore types based on morphological parameters. The technical solution of this invention overcomes the problem of inaccurate and rapid segmentation of organic pores and fractures in shale MAPs in existing technologies.

Description

Technical Field

[0001] This invention relates to the fields of oil and gas development and image segmentation, specifically to a rapid identification method for organic pores and fractures in shale MAPs based on FCM. Background Technology

[0002] As an unconventional oil and gas reservoir, shale reservoirs exhibit diverse pore types and complex pore structures, including nanoscale organic pores and fractures, as well as nanoscale inorganic pores and fractures. Accurate identification and characterization of organic pores and organic fractures are fundamental to describing the storage and permeability of shale reservoirs. With the development of computer and imaging technologies, macroscopically oriented pore structures (MAPs) have been used to characterize the multi-scale pore structure of shale reservoirs. Given the large number of nanoscale organic and inorganic pores and fractures in MAPs, distinguishing between pore types and pore morphologies is necessary to accurately characterize the storage and permeability of shale reservoirs.

[0003] Gray-scale-based conventional segmentation methods are among the earliest and simplest techniques used in shale MAP porosity extraction. Gray-scale segmentation distinguishes pores, minerals, and organic matter based on their gray-scale differences. Typical methods include the Otsu's method, the watershed algorithm, and threshold-based region growing methods. These methods are computationally efficient, require few parameters, and can quickly achieve preliminary pore identification in images with clear gray-scale contrast and low noise. However, shale MAP images often suffer from severe gray-scale overlap and similar mineral gray-scale values, making it difficult for a single gray-scale threshold to accurately distinguish different pore types. Furthermore, this method is highly sensitive to noise and local extrema, easily leading to oversegmentation and requiring complex pre- and post-processing steps. Therefore, gray-scale-based conventional segmentation methods are more suitable as preliminary segmentation or auxiliary steps, and their applicability in high-precision quantitative porosity analysis is significantly limited.

[0004] Neural network-based pore segmentation methods learn the nonlinear relationship between pores and matrix in images, demonstrating strong adaptability under complex grayscale and texture conditions. Convolutional Neural Networks (CNNs) can automatically extract multi-scale spatial features, significantly outperforming traditional manual feature extraction methods; Generative Adversarial Networks (GANs), through adversarial training mechanisms, possess unique advantages in pore boundary enhancement, few-shot learning, and data augmentation. These methods effectively mitigate grayscale overlap and noise interference, achieving high segmentation accuracy in complex shale MAP images. However, their performance is highly dependent on high-quality labeled data and the number of training samples, while manual labeling of pore boundaries is inherently subjective and uncertain. Furthermore, neural network models typically lack explicit physical constraints, resulting in relatively weak interpretability of segmentation results. Therefore, while neural network-based methods show high potential in pore segmentation, further optimization combining geological priors and physical constraints is still necessary.

[0005] Deep learning-based segmentation methods have become a research hotspot in shale MAP (mapped mineral structures) pore extraction in recent years. A key characteristic is the use of an end-to-end semantic segmentation framework, directly outputting pixel-level segmentation results from the original image. Represented by models such as U-Net and DeepLab, these methods effectively improve the ability to identify micropores and complex boundaries by fusing encoder-decoder structures with multi-scale features. Deep learning methods can simultaneously consider grayscale, texture, and spatial context information in shale multi-mineral systems, significantly improving the overall consistency and stability of pore segmentation. However, their training process requires significant computational resources, and the model's generalization ability is easily affected by imaging conditions, resolution, and lithological differences.

[0006] Hard clustering-based pore segmentation methods, typically represented by K-means, forcibly divide image pixels into several non-overlapping categories based on grayscale or feature distance. These methods are simple to implement, computationally efficient, and do not rely on manually labeled data, thus avoiding the influence of subjective factors to some extent. Therefore, they have found some application in the preliminary identification of shale pores. However, hard clustering methods assume that each pixel belongs to only a single category, making it difficult to describe the pore-fracture transition zones and grayscale mixing zones commonly found in shale. Furthermore, the number of clusters needs to be pre-set manually and is sensitive to the initial cluster centers, easily leading to unstable results. For shale MAPs images with complex pore structures and multiphase characteristics, hard clustering methods often fail to accurately reflect the true pore spatial structure.

[0007] Therefore, there is a need for a fast identification method for organic pores and fractures in shale MAPs based on FCM that can accurately and quickly segment the pores of shale MAPs. Summary of the Invention

[0008] The main objective of this invention is to provide a rapid identification method for organic pores and fractures in shale MAPs based on FCM, in order to solve the problem that existing technologies cannot accurately and quickly segment the pores of shale MAPs.

[0009] To achieve the above objectives, this invention provides a rapid identification method for organic pores and fractures in shale MAPs based on FCM, specifically including the following steps: S1, a self-guided filtering preprocessing is performed on each image block, and the filtered image blocks are finally combined into complete MAPs.

[0010] S2, segmenting the filtered MAPs based on the FCM algorithm.

[0011] S3 uses the average gray value of the neighborhood of the pore to determine the pore type.

[0012] S4, pore type differentiation based on morphological parameters.

[0013] Furthermore, step S1 specifically includes the following steps: S1.1, using the Python extension library tifffile to read shale MAPs in TIFF format into a Python program to obtain images. Number of pixels in direction and Number of pixels in direction At this time, the MAPs image is a shape of matrix .

[0014] S1.2, the image block size in block filtering is set to... The overlapping area between pixels and image blocks is Pixels, regularization parameter is Guide image For the image patch itself; guided filtering is implemented using the boxFilter function of the Python extension library OpenCV, and the boxFilter is called cyclically in each image patch using self-written code.

[0015] S1.3, use the Python extension library numpy to create a matrix of the same size as the original MAPs. The initial values ​​are all 0. The image after block filtering is added one by one. In the middle, use OpenCV to save, This is the filtered image of the MAPs.

[0016] Furthermore, step S2 specifically includes the following steps: S2.1, Define an objective function, and randomly select... We first generate an initial cluster center, then assign each sample to a specific cluster based on its distance from the cluster center, and finally recalculate the cluster center for each cluster. FCM objective function The common form is: ; in, The membership matrix, For the set of cluster centers, For MAPs (pixels), Indicates the first The pixel belongs to the first The membership degree of a cluster satisfies , For fuzzy weighted index, For the first A vector of pixels, For the first Cluster centers; The constraints are: ; The iterative update formula for cluster centers is obtained by finding the extreme value of the objective function: .

[0017] S2.2, Repeat step S2.1, continuously updating the membership matrix and cluster centers until the objective function converges to its minimum value or reaches the preset maximum number of iterations, finally obtaining the clustering result, and converting the clustering result into a binary pore image. .

[0018] Furthermore, step S3 specifically includes the following steps: S3.1, based on The location of development in MAPs, i.e., two-dimensional spatial coordinates, will Morphological dilation is performed using a radius of... The circular structure expands outward, resulting in the expanded region. Morphological dilatation is represented as: ; in, For radius Structural elements; It is a symbol for morphological expansion.

[0019] right and Perform Boolean subtraction to obtain neighborhood Boolean subtraction is represented as: .

[0020] S3.2, Obtain the pore neighborhood Afterwards, statistics Mean grayscale value: ; in, This represents the average gray value of the neighborhood of the pores; This represents the number of pixels within that neighborhood region. Let be the coordinates of a point in that neighborhood; for The grayscale value.

[0021] S3.3, set the organic matter grayscale threshold to When the average gray value of the pore neighborhood is less than or equal to When the average gray value of the pore neighborhood is greater than a certain value, the pore is identified as an organic matter pore; when the average gray value of the pore neighborhood is greater than a certain value, the pore is identified as an organic matter pore. At that time, the pore was identified as an inorganic pore, expressed by the formula: .

[0022] Furthermore, step S4 specifically includes the following steps: S4.1, using eccentricity to describe connected regions, the expression for eccentricity is: ; ; in, The eccentricity of a single hole ranges from [0, 1]. The semi-major axis of the equivalent ellipse, in pixels; The semi-minor axis of the equivalent ellipse is expressed in pixels.

[0023] S4.2, using aspect ratio to distinguish between pores and cracks, the expression is: ; ; in, The aspect ratio of a single pore. ; and These are the major and minor axes of the equivalent ellipse of the pore, respectively.

[0024] Furthermore, step S4 also includes the following steps: S4.3, obtained based on morphological methods Calculate the equivalent ellipse for each hole. and ;when Values ​​greater than 0.9~0.995 and When the value is not less than 4, the hole morphology is a crack, and it is labeled F; otherwise... and In the combination of these elements, the pore shape is a pore, labeled P, and the result after adding the label... Stored as This achieves the identification of organic pores and fractures in shale MAPs, and OpenCV is used to process the images. storage.

[0025] S4.4, Count the number of pixels for organic holes and organic seams. , Then the porosity of organic pores Organic seam surface rate Organic pore ratio Organic seam ratio The fraction is represented as: ; ; ; ; in, The number of pixels in the binarized pore image of shale MAPs. .

[0026] The present invention has the following beneficial effects: This invention utilizes block-guided filtering to accelerate noise removal, employs the FCM algorithm for pore segmentation based on MAP grayscale, and combines the mean grayscale value of the neighborhood with pore morphology parameters to extract organic pores and organic fractures. This enables rapid quantitative characterization of organic pores and fractures in shale MAPs, which is the foundation for evaluating shale reservoir and seepage capacity. Attached Figure Description

[0027] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart of a rapid identification method for organic pores and fractures in shale MAPs based on FCM is shown.

[0028] Figure 2 A map of shale MAPs is shown.

[0029] Figure 3 The image shows the mineral and porosity composition of unfiltered shale MAPs.

[0030] Figure 4 The image shows a block of filtered shale MAPs.

[0031] Figure 5 The image shows a pore segmentation image of a block of shale MAPs.

[0032] Figure 6 A schematic diagram illustrating the principle of obtaining a single pore neighborhood through morphological dilation is shown.

[0033] Figure 7 Images of pores of different types and morphologies are shown. Detailed Implementation

[0034] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] like Figure 1 The method for rapid identification of organic pores and fractures in shale MAPs based on FCM, as shown, specifically includes the following steps: S1, a self-guided filtering preprocessing is performed on each image block, and the filtered image blocks are finally combined into complete MAPs.

[0036] S2, segmenting the filtered MAPs based on the FCM algorithm.

[0037] S3 uses the average gray value of the neighborhood of the pore to determine the pore type.

[0038] S4, pore type differentiation based on morphological parameters.

[0039] Specifically, due to the massive data volume of MAPs, direct filtering based on MAPs consumes enormous computational resources. Therefore, a MAP block filtering strategy is proposed, employing guided filtering preprocessing for each image block, which can better preserve the edge features of apertures. Finally, the filtered image blocks are combined into complete MAPs, thus completing the MAP preprocessing. Step S1 specifically includes the following steps: S1.1, using the Python extension library tifffile to read shale MAPs in TIFF format into a Python program to obtain images. Number of pixels in direction and Number of pixels in direction At this time, the MAPs image is a shape of matrix .like Figure 2 As shown, the number of pixels Nx in the X direction of the image is 36864, and the number of pixels Ny in the Y direction is 36864. At this time, the MAPs image is a matrix with a shape of [36864, 36864]. .

[0040] S1.2, the image block size in block filtering is set to... The overlapping area between pixels and image blocks is Pixels, regularization parameter is Guide image The guided filter is the image patch itself. The guided filter is implemented using the boxFilter function of the Python extension library OpenCV. The boxFilter is called in a loop in each image patch using self-written code. The input parameters of the boxFilter function are set as follows: the image patch size is set to 1024 pixels, the overlap area between patches is 15 pixels, the regularization parameter value is 0.001×255×255, and the guided image is the image patch itself, i.e., the guided filter.

[0041] Because MAPs images have a huge number of pixels, the complete image cannot display details. (Example image patch) Its guiding image is ,like Figure 3 As shown.

[0042] S1.3, use the Python extension library numpy to create a matrix of the same size as the original MAPs. The initial values ​​are all 0. The image after block filtering is added one by one. In the middle, use OpenCV to save, This is the filtered MAPs image. (Filtered image patch) like Figure 4 As shown.

[0043] Specifically, step S2 includes the following steps: S2.1, for When performing FCM clustering, it is necessary to define an objective function and randomly select... We first generate an initial cluster center, then assign each sample to a specific cluster based on its distance from the cluster center, and finally recalculate the cluster center for each cluster. FCM objective function The common form is: ; in, The membership matrix, For the set of cluster centers, For MAPs (pixels), Indicates the first The pixel belongs to the first The membership degree of a cluster satisfies , It is a fuzzy weighted index (usually with a value of 2). For the first A vector of pixels, For the first Cluster centers.

[0044] The constraints are: ; The iterative update formula for cluster centers is obtained by finding the extreme value of the objective function: .

[0045] S2.2, Repeat step S2.1, continuously updating the membership matrix and cluster centers until the objective function converges to its minimum value or reaches the preset maximum number of iterations, finally obtaining the clustering result, and converting the clustering result into a binary pore image. .

[0046] The FCM algorithm is implemented using the `fuzz.cluster.cmeans()` function from the Python extension library `skfuzzy`. Since it does not consider distinguishing between various minerals in shale MAPs, the number of clusters is low. Setting it to 2 means only image segmentation of pores and minerals is performed. The input image is... Fuzziness Index The threshold value is 1.8. The maximum number of iterations for the FCM algorithm is typically set to 1000 steps, with a value of 0.005, to prevent the program from getting stuck in infinite iteration.

[0047] The process of segmenting MAPs using the FCM algorithm is as follows: First, input the parameters mentioned above and select 2 clusters, meaning there are 2 initial cluster centers. Then, based on the distance of each pixel value to the cluster center, assign the pixel to a specific cluster. Finally, recalculate the cluster center for each cluster. This process is repeated until a termination threshold is reached.

[0048] After segmentation using the FCM algorithm, a binarized pore image is obtained. Save using OpenCV This is used for subsequent identification of pore types. Segmented image like Figure 5 As shown.

[0049] Specifically, after obtaining the binarized pore image of shale MAPs, the pore type can be further distinguished based on the image, using the average gray value of the pore's neighborhood to determine the pore type. The specific principle is as follows: the gray values ​​of shale MAPs represent different attributes, with a gray value range of [0, 255]. The lower the gray value, the darker the image, and vice versa. Among them, pores have the lowest gray value (i.e., the darkest image), followed by organic matter, then minerals such as quartz, feldspar, and clay, and pyrite has the highest gray value (i.e., the brightest image). Therefore, it can be determined whether the pore is developed within organic matter based on the average gray value of the pore's neighborhood. If the average gray value of the pore's neighborhood is low (usually below 70), that is, the pore is developed in organic matter, and the pore is identified as an organic matter pore. Step S3 specifically includes the following steps: S3.1, as Figure 6 As shown, the neighborhood of the pore image is obtained: based on The location of development in MAPs, i.e., two-dimensional spatial coordinates, will Morphological dilation is performed using a radius of... The circular structure expands outward, resulting in the expanded region. Morphological dilatation is represented as: ; in, For radius The structural element, without units; It is a symbol for morphological expansion.

[0050] right and Perform Boolean subtraction to obtain neighborhood Boolean subtraction is represented as: .

[0051] S3.2, Calculate the average gray value of the pore neighborhood. Obtain the pore neighborhood. Afterwards, statistics Mean grayscale value: ; in, This represents the average gray value of the pore neighborhood region; it has no unit. This represents the number of pixels within the neighborhood, without units. The coordinates of a point in this neighborhood are unitless. for grayscale value, The range is [0, 255], and it has no units.

[0052] S3.3, Pore type determination: Set the organic matter grayscale threshold to... When the average gray value of the pore neighborhood is less than or equal to When the average gray value of the pore neighborhood is greater than a certain value, the pore is identified as an organic matter pore; when the average gray value of the pore neighborhood is greater than a certain value, the pore is identified as an organic matter pore. At that time, the pore was identified as an inorganic pore, expressed by the formula: .

[0053] The grayscale threshold for organic matter is set to 70. When the average grayscale value G of the pore neighborhood is less than or equal to 70, the pore is identified as an organic matter pore; when G is greater than 70, the pore is identified as an inorganic matter pore. That is: .

[0054] By assigning labels OM to organic matter pores and IOM to inorganic matter pores, different pore types can be distinguished in the binary pore images of shale MAPs.

[0055] Specifically, based on morphological parameters, the differentiation of organic pores and fractures is crucial. For organic pores of interest to technical personnel, further differentiation of pore morphology (pores and fractures) is needed to evaluate the storage and seepage capacity of shale organic pores. A high proportion of pores indicates better shale storage capacity, while a high proportion of fractures indicates better seepage capacity. Pore morphology is determined by combining eccentricity and aspect ratio.

[0056] Step S4 specifically includes the following steps: S4.1, the eccentricity is used to describe the connected region, that is, the degree to which each pore in the binary pore image deviates from the circle. The expression for eccentricity is: ; ; in, The eccentricity of a single pore, ranging from [0, 1], is defined by e = 0.9, which can distinguish between pores and cracks to some extent. It has no unit. The semi-major axis of the equivalent ellipse, in pixels; The semi-minor axis of the equivalent ellipse is expressed in pixels.

[0057] S4.2, Aspect ratio is one of the most intuitive and commonly used morphological indicators for distinguishing between porosity and cracks. The expression for using aspect ratio to distinguish between porosity and cracks is: ; ; in, The aspect ratio of a single pore. No unit; and These are the major and minor axes of the equivalent ellipse of the pore, respectively, in pixels.

[0058] Specifically, step S4 also includes the following steps: S4.3, obtained based on morphological methods Calculate the equivalent ellipse for each hole. and ;when Values ​​greater than 0.9~0.995 and When the value is not less than 4, the hole morphology is a crack, and it is labeled F; otherwise... and In the combination of these elements, the pore shape is a pore, labeled P, and the result after adding the label... Stored as This achieves the identification of organic pores and fractures in shale MAPs, and OpenCV is used to process the images. Storage. For example image patch Its aperture segmentation image patch Images of different types and morphologies of pores after identification. ,like Figure 7 As shown.

[0059] S4.4 introduces face rate , representing the two-dimensional porosity of shale MAPs, i.e., the proportion of pores in MAPs, is expressed as: .

[0060] Count the number of pixels for organic holes (labeled OM-P) and organic slits (labeled OM-F). , Then the porosity of organic pores Organic seam surface rate Organic pore ratio Organic seam ratio The fraction is represented as: ; ; ; ; in, The number of pixels in the binarized pore image of shale MAPs. No unit.

[0061] calculate Figure 2 The characterization parameters of the MAPs shown in Table 1 are as follows:

[0062] Table 1 Characterization parameters of MAPs

[0063] Data shows that the open pore rate of these MAPs is low, only 0.578%. The total percentage of organic pores ( + The organic porosity was 10.3%, indicating a high proportion of organic pores. The organic porosity was more than five times that of organic fractures, meaning it had a stronger oil and gas storage capacity.

[0064] This invention provides a rapid identification method for organic pores and fractures in shale MAPs based on FCM. By using block-guided filtering, the filtering process for large-volume MAPs is reduced. The pore segmentation based on the FCM algorithm better handles the ambiguity of pore boundaries, and the identification of organic pores and fractures is achieved by using the gray-scale mean of neighboring pixels and morphological parameters, thereby realizing the quantitative characterization of organic pores and fractures and providing technical support for evaluating the storage and seepage capacity of shale reservoirs.

[0065] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A rapid identification method for organic pores and fractures in shale MAPs based on FCM, characterized in that, Specifically, the steps include the following: S1, a self-guided filtering preprocessing is performed on each image block, and the filtered image blocks are finally combined into complete MAPs; S2, segmenting the filtered MAPs based on the FCM algorithm; S3, the pore type is determined by the average gray value of the pore's neighborhood; S4, pore type differentiation based on morphological parameters.

2. The method for rapid identification of organic pores and fractures in shale MAPs based on FCM according to claim 1, characterized in that, Step S1 specifically includes the following steps: S1.1, using the Python extension library tifffile to read shale MAPs in TIFF format into a Python program to obtain images. Number of pixels in direction and Number of pixels in direction At this time, the MAPs image is a shape of matrix ; S1.2, the image block size in block filtering is set to... The overlapping area between pixels and image blocks is Pixels, regularization parameter is Guide image For the image patch itself; guided filtering is implemented using the boxFilter function of the Python extension library OpenCV, and the boxFilter is called cyclically in each image patch using self-written code; S1.3, use the Python extension library numpy to create a matrix of the same size as the original MAPs. The initial values ​​are all 0. The image after block filtering is added one by one. In the middle, use OpenCV to save, This is the filtered image of the MAPs.

3. The method for rapid identification of organic pores and fractures in shale MAPs based on FCM according to claim 1, characterized in that, Step S2 specifically includes the following steps: S2.1, Define an objective function, and randomly select... We first generate an initial cluster center, then assign each sample to a specific cluster based on its distance from the cluster center, and finally recalculate the cluster center for each cluster. FCM objective function The common form is: ； in, The membership matrix, For the set of cluster centers, For MAPs (pixels), Indicates the first The pixel belongs to the first The membership degree of a cluster satisfies , For fuzzy weighted index, For the first A vector of pixels, For the first Cluster centers; The constraints are: ； The iterative update formula for cluster centers is obtained by finding the extreme value of the objective function: ； S2.2, Repeat step S2.1, continuously updating the membership matrix and cluster centers until the objective function converges to its minimum value or reaches the preset maximum number of iterations, finally obtaining the clustering result, and converting the clustering result into a binary pore image. .

4. The method for rapid identification of organic pores and fractures in shale MAPs based on FCM according to claim 1, characterized in that, Step S3 specifically includes the following steps: S3.1, based on The location of development in MAPs, i.e., two-dimensional spatial coordinates, will Morphological dilation is performed using a radius of... The circular structure expands outward, resulting in the expanded region. Morphological dilatation is represented as: ； in, For radius Structural elements; It is a morphological expansion symbol; right and Perform Boolean subtraction to obtain neighborhood Boolean subtraction is represented as: ； S3.2, Obtain the pore neighborhood Afterwards, statistics Mean grayscale value: ； in, This represents the average gray value of the neighborhood of the pores; This represents the number of pixels within that neighborhood region. Let be the coordinates of a point in that neighborhood; for grayscale value; S3.3, set the organic matter grayscale threshold to When the average gray value of the pore neighborhood is less than or equal to When the average gray value of the pore neighborhood is greater than a certain value, the pore is identified as an organic matter pore; when the average gray value of the pore neighborhood is greater than a certain value, the pore is identified as an organic matter pore. At that time, the pore was identified as an inorganic pore, expressed by the formula: 。 5. The method for rapid identification of organic pores and fractures in shale MAPs based on FCM according to claim 1, characterized in that, Step S4 specifically includes the following steps: S4.1, using eccentricity to describe connected regions, the expression for eccentricity is: ； ； in, The eccentricity of a single hole ranges from [0, 1]. The semi-major axis of the equivalent ellipse, in pixels; The semi-minor axis of the equivalent ellipse is expressed in pixels. S4.2, using aspect ratio to distinguish between pores and cracks, the expression is: ； ； in, The aspect ratio of a single pore. ; and These are the major and minor axes of the equivalent ellipse of the pore, respectively.

6. The method for rapid identification of organic pores and fractures in shale MAPs based on FCM according to claim 5, characterized in that, Step S4 also includes the following steps: S4.3, obtained based on morphological methods Calculate the equivalent ellipse for each hole. and ;when Values ​​greater than 0.9~0.995 and When the value is not less than 4, the hole morphology is a crack, and it is labeled F; otherwise... and In the combination of these elements, the pore shape is a pore, labeled P, and the result after adding the label... Stored as This achieves the identification of organic pores and fractures in shale MAPs, and OpenCV is used to process the images. storage; S4.4, Count the number of pixels for organic holes and organic seams. , Then the porosity of organic pores Organic seam surface rate Organic pore ratio Organic seam ratio The fraction is represented as: ； ； ； ； in, The number of pixels in the binarized pore image of shale MAPs. .

Images